Source author record

Chang Liu

Chang Liu appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Computer Vision astro-ph.SR Machine Learning cond-mat.mes-hall cond-mat.mtrl-sci cond-mat.str-el Artificial Intelligence cond-mat.supr-con eess.IV gr-qc Computation and Language Cryptography and Security Robotics eess.SY Networking and Internet Architecture Systems and Control astro-ph.IM eess.SP physics.comp-ph physics.plasm-ph Multiagent Systems quant-ph astro-ph.HE Information Theory math.IT math.NA physics.app-ph physics.flu-dyn eess.AS Information Retrieval math.OC Numerical Analysis physics.chem-ph Sound Computational Engineering, Finance, and Science Graphics hep-th math.AP math.CO Neural and Evolutionary Computing physics.data-an physics.optics astro-ph.CO astro-ph.GA Biomolecules cond-mat.dis-nn cond-mat.other cond-mat.quant-gas cond-mat.soft cond-mat.stat-mech cs.CY Data Structures and Algorithms econ.TH hep-ex hep-ph math.PR math.RA math.ST Methodology Multimedia Neurons and Cognition Quantitative Methods Statistics Theory

Catalog footprint

What is connected

277works

63topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

3D Dynamic Radio Map Prediction Using Vision Transformers for Low-Altitude Wireless Networks

Low-altitude wireless networks (LAWN) are rapidly expanding with the growing deployment of unmanned aerial vehicles (UAVs) for logistics, surveillance, and emergency response. Reliable connectivity remains a critical yet challenging task due to three-dimensional (3D) mobility, time-varying user density, and limited power budgets. The transmit power of base stations (BSs) fluctuates dynamically according to user locations and traffic demands, leading to a highly non-stationary 3D radio environment. Radio maps (RMs) have emerged as an effective means to characterize spatial power distributions and support radio-aware network optimization. However, most existing works construct static or offline RMs, overlooking real-time power variations and spatio-temporal dependencies in multi-UAV networks. To overcome this limitation, we propose a 3D dynamic radio map (3D-DRM) framework that learns and predicts the spatio-temporal evolution of received power. Specially, a Vision Transformer (ViT) encoder extracts high-dimensional spatial representations from 3D RMs, while a Transformer-based module models sequential dependencies to predict future power distributions. Experiments unveil that 3D-DRM accurately captures fast-varying power dynamics and substantially outperforms baseline models in both RM reconstruction and short-term prediction.

preprint2026arXiv

Beyond Factor Aggregation: Gauge-Aware Low-Rank Server Representations for Federated LoRA

Federated LoRA enables parameter-efficient adaptation of large language models under decentralized data and limited client resources.However, directly averaging LoRA factors is representation-dependent: the same intrinsic update admits infinitely many gauge-equivalent factorizations, so factor-level aggregation can change under arbitrary coordinate choices while the underlying update remains unchanged. This reveals a semantic mismatch in existing federated LoRA aggregation rules. We propose \textbf{GLoRA}, a gauge-aware server representation for federated LoRA.Instead of aggregating raw factors, GLoRA estimates a consensus update subspace from client projectors and aggregates client updates in shared reference coordinates, thereby representing semantic update aggregation entirely in low-rank form. To support heterogeneous client capacities, GLoRA further provides a rank-compatible readout that instantiates adapters of different ranks from the same server state without dense update reconstruction. Experiments on GLUE and SuperNI show that GLoRA consistently outperforms federated LoRA baselines under data, resource, and task heterogeneity, including heterogeneous client ranks, sparse participation, larger backbones, and unseen-task evaluation. GLoRA also achieves a favorable efficiency--performance trade-off, suggesting that effective federated LoRA requires not merely averaging low-rank factors, but defining a semantically meaningful server-side representation for aggregation.

preprint2026arXiv

Biosignal Fingerprinting: A Cross-Modal PPG-ECG Foundation Model

Cardiovascular disease remains the leading cause of global mortality, yet scalable cardiac monitoring is hindered by the gap between diagnostic-rich ECG and ubiquitous wearable PPG. Bridging this gap requires representations that are compact, transferable across modalities and devices, and deployable without task-specific retraining. Here we introduce biosignal fingerprints: compact latent representations of cardiovascular state derived from a cross-modal foundation model, the Multi-modal Masked Autoencoder (M2AE), trained on over 3.4 million paired ECG and PPG signals. M2AE integrates modality-specific encoders with a shared bottleneck and dual decoders, jointly optimized using reconstruction and cross-modal contrastive objectives, yielding generalizable fingerprints that retain intra- and inter-modality features. Like a biometric fingerprint, these representations uniquely encode an individual's cardiovascular state in a modality-agnostic, privacy-preserving form reusable across clinical tasks without exposing raw waveform data or requiring model retraining. Across 7 downstream tasks, spanning cross-modal reconstruction, cardiovascular disease classification, hypertension detection, mortality prediction, and demographic inference, biosignal fingerprints achieve competitive or superior performance compared to leading domain-specialist foundation models in frozen settings, including an AUROC of 0.974 for five-class CVD classification and 0.877 for hypertension detection, with a maximum improvement of 27.7% in AUROC across 5 classification tasks. Critically, strong performance is maintained with only a single modality, enabling deployment in resource-constrained, single-sensor environments typical of real-world wearable monitoring, with direct implications for continuous cardiovascular monitoring across clinical and consumer health settings.

preprint2026arXiv

Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment

The aesthetic quality assessment task is crucial for developing a human-aligned quantitative evaluation system for AIGC. However, its inherently complex nature, spanning visual perception, cognition, and emotion, poses fundamental challenges. Although aesthetic descriptions offer a viable representation of this complexity, two critical challenges persist: (1) data scarcity and imbalance: existing dataset overly focuses on visual perception and neglects deeper dimensions due to the expensive manual annotation; and (2) model fragmentation: current visual networks isolate aesthetic attributes with multi-branch encoder, while multimodal methods represented by contrastive learning struggle to effectively process long-form textual descriptions. To resolve challenge (1), we first present the Refined Aesthetic Description (RAD) dataset, a large-scale (70k), multi-dimensional structured dataset, generated via an iterative pipeline without heavy annotation costs and easy to scale. To address challenge (2), we propose ArtQuant, an aesthetics assessment framework for artistic images which not only couples isolated aesthetic dimensions through joint description generation, but also better models long-text semantics with the help of LLM decoders. Besides, theoretical analysis confirms this symbiosis: RAD's semantic adequacy (data) and generation paradigm (model) collectively minimize prediction entropy, providing mathematical grounding for the framework. Our approach achieves state-of-the-art performance on several datasets while requiring only 33% of conventional training epochs, narrowing the cognitive gap between artistic images and aesthetic judgment. We will release both code and dataset to support future research.

preprint2026arXiv

Bridging Distance and Spectral Positional Encodings via Anchor-Based Diffusion Geometry Approximation

Molecular graph learning benefits from positional signals that capture both local neighborhoods and global topology. Two widely used families are spectral encodings derived from Laplacian or diffusion operators and anchor-based distance encodings built from shortest-path information, yet their precise relationship is poorly understood. We interpret distance encodings as a low-rank surrogate of diffusion geometry and derive an explicit trilateration map that reconstructs truncated diffusion coordinates from transformed anchor distances and anchor spectral positions, with pointwise and Frobenius-gap guarantees on random regular graphs. On DrugBank molecular graphs using a shared GNP-based DDI prediction backbone, a distance-driven Nyström scheme closely recovers diffusion geometry, and both Laplacian and distance encodings substantially outperform a no-encoding baseline.

preprint2026arXiv

Chain-based Distillation for Effective Initialization of Variable-Sized Small Language Models

Large language models (LLMs) achieve strong performance but remain costly to deploy in resource-constrained settings. Training small language models (SLMs) from scratch is computationally expensive, while conventional knowledge distillation requires repeated access to large teachers for different target sizes, leading to poor scalability. To solve these problems, we propose \textbf{Chain-based Distillation (CBD)}, a scalable paradigm for efficiently initializing variable-sized language models. A sparse and limited sequence of intermediate models (called anchors) is constructed via stepwise distillation, forming a distillation chain that progressively transfers knowledge from the source LLMs. To support heterogeneous settings, we introduce \emph{bridge distillation} for cross-architecture and cross-vocabulary transfer. Models of variable sizes are initialized via parameter interpolation between adjacent anchors, eliminating repeated large teacher inference. Experiments show that the proposed method substantially improves efficiency and downstream performance. A 138M-parameter SLM without recovery pre-training, outperforms scratch-trained models on a 10B-token corpus on the specific task. CBD also demonstrates versatility in heterogeneous settings for initialize models with different architectures and vocabularies.

preprint2026arXiv

CoINS: Counterfactual Interactive Navigation via Skill-Aware VLM

Recent Vision-Language Models (VLMs) have demonstrated significant potential in robotic planning. However, they typically function as semantic reasoners, lacking an intrinsic understanding of the specific robot's physical capabilities. This limitation is particularly critical in interactive navigation, where robots must actively modify cluttered environments to create traversable paths. Existing VLM-based navigators are predominantly confined to passive obstacle avoidance, failing to reason about when and how to interact with objects to clear blocked paths. To bridge this gap, we propose Counterfactual Interactive Navigation via Skill-aware VLM (CoINS), a hierarchical framework that integrates skill-aware reasoning and robust low-level execution. Specifically, we fine-tune a VLM, named InterNav-VLM, which incorporates skill affordance and concrete constraint parameters into the input context and grounds them into a metric-scale environmental representation. By internalizing the logic of counterfactual reasoning through fine-tuning on the proposed InterNav dataset, the model learns to implicitly evaluate the causal effects of object removal on navigation connectivity, thereby determining interaction necessity and target selection. To execute the generated high-level plans, we develop a comprehensive skill library through reinforcement learning, specifically introducing traversability-oriented strategies to manipulate diverse objects for path clearance. A systematic benchmark in Isaac Sim is proposed to evaluate both the reasoning and execution aspects of interactive navigation. Extensive simulations and real-world experiments demonstrate that CoINS significantly outperforms representative baselines, achieving a 17\% higher overall success rate and over 80\% improvement in complex long-horizon scenarios compared to the best-performing baseline

preprint2026arXiv

Count Anything at Any Granularity

Open-world object counting remains brittle: despite rapid advances in vision-language models (VLMs), reliably counting the objects a user intends is far from solved. We argue that a central reason is that counting granularity is left implicit; users may refer to a specific identity, an attribute, an instance type, a category, or an abstract concept, yet most methods treat "what to count" as a single, category-level matching problem. In this work, we redefine open-world counting as multi-grained counting, where visual exemplars specify target appearance and fine-grained text, with optional negative prompts, specifies the intended semantic granularity across five explicit levels. Making granularity explicit, however, exposes a critical data bottleneck: existing counting datasets lack the multi-category scenes, controlled distractors, and instance-level annotations needed to verify fine-grained prompt semantics. To address this, we propose the first fully automatic data-scaling pipeline that integrates controllable 3D synthesis with consistent image editing and VLM-based filtering, and use it to construct KubriCount, the largest and most comprehensively annotated counting dataset to date, supporting both training and multi-grained evaluation. Systematic benchmarking reveals that both multimodal large language models and specialist counting models exhibit severe prompt-following failures under fine-grained distinctions. Motivated by these findings, we train HieraCount, a multi-grained counting model that jointly leverages text and visual exemplars as complementary target specifications. HieraCount substantially improves multi-grained counting accuracy and generalizes robustly to challenging real-world scenarios. The project page is available here: https://verg-avesta.github.io/KubriCount/.

preprint2026arXiv

DCGL: Dual-Channel Graph Learning with Large Language Models for Knowledge-Aware Recommendation

Knowledge Graphs (KGs) have proven highly effective for recommendation systems by capturing latent item relationships, while recent integration of Large Language Models (LLMs) has further enhanced semantic understanding and addressed knowledge sparsity issues. Nevertheless, current KG-and-LLM-based methods still face three main limitations: 1) inadequate modeling of implicit semantic relationships beyond explicit KG links; 2) suboptimal single-channel fusion of ID and LLM embeddings, which often leads to signal interference and blurred representations; and 3) insufficient consideration of user-item interaction frequency variations in recommendation strategies. To address these challenges, we propose the Dual-Channel Graph Learning (DCGL) framework, featuring three key innovations: 1) a dual-channel architecture that structurally decouples rich semantic information from user behavioral patterns, preventing early interference; 2) a multi-level contrastive learning mechanism that enhances robustness against KG noise through intra-view contrasts and bridges semantic gaps between channels via inter-view alignment; and 3) a dynamic fusion mechanism that adaptively balances semantic generalization and behavioral specificity based on interaction frequency, resolving the cascading limitation. Extensive experiments on four real-world datasets show that DCGL consistently outperforms state-of-the-art methods, yielding substantial improvements in sparse scenarios while maintaining precision for active users. Our code is available at https://github.com/XinchiZou/DCGL.

preprint2026arXiv

Diffusion Model-Enhanced Environment Reconstruction in ISAC

Recently, environment reconstruction (ER) in integrated sensing and communication (ISAC) systems has emerged as a promising approach for achieving high-resolution environmental perception. However, the initial results obtained from ISAC systems are coarse and often unsatisfactory due to the high sparsity of the point clouds and significant noise variance. To address this problem, we propose a noise-sparsity-aware diffusion model (NSADM) post-processing framework. Leveraging the powerful data recovery capabilities of diffusion models, the proposed scheme exploits spatial features and the additive nature of noise to enhance point cloud density and denoise the initial input. Simulation results demonstrate that the proposed method significantly outperforms existing model-based and deep learning-based approaches in terms of Chamfer distance and root mean square error.

preprint2026arXiv

Dynamic Water-Wave Tweezers

Following a recent demonstration of stable trapping of floating particles by stationary (monochromatic) structured water waves [Nature 638, 394 (2025)], we report dynamic water-wave tweezers that enable controllable transport of trapped particles along arbitrary trajectories on the water surface. We employ a triangular lattice formed by the interference of three plane waves, which can trap particles, depending on parameters, either at intensity maxima or at intensity zeros (vortices). By introducing small frequency detunings between the interfering waves, we control 2D motion of the lattice and trapped particles. This approach is robust and effective over a relatively broad range of particle sizes and wave frequencies, offering remarkable new possibilities for noncontact manipulation of floating (e.g., biological and soft-matter) objects in fluidic environments.

preprint2026arXiv

EgoIntrospect: An Egocentric Dataset and Benchmark for User-Centric Internal State Reasoning

Despite extensive efforts on egocentric video datasets and benchmarks, understanding users' internal states, which is crucial for enabling seamless AI assistant experiences, remains largely overlooked. In this work, we introduce EgoIntrospect, the first egocentric dataset captured in user-driven scenarios with self-annotations that explicitly reveal users' interactive intentions with AI assistants. EgoIntrospect was collected using a cross-device setup, providing synchronized video, audio, gaze, motion, and physiological signals. It consists of 180 hours of recordings from 60 subjects, with an average recording duration of 3 hours per subject. Leveraging EgoIntrospect, we formalize a suite of tasks centered on user internal states, including affective experience, interactive intent, and cognitive memory. We further process the annotations to construct benchmarks that evaluate the ability of modern multimodal large language models to reason about users' internal states from egocentric observations. Experiments on our benchmark suggest that existing multimodal large language models struggle to effectively leverage multimodal signals to infer users' subjective internal states. The dataset and annotations will be made publicly available to advance research in egocentric vision and wearable AI assistants. Project page: https://ego-introspect.github.io/

preprint2026arXiv

FinDeepForecast: A Live Multi-Agent System for Benchmarking Deep Research Agents in Financial Forecasting

Deep Research (DR) Agents powered by advanced Large Language Models (LLMs) have fundamentally shifted the paradigm for completing complex research tasks. Yet, a comprehensive and live evaluation of their forecasting performance on real-world, research-oriented tasks in high-stakes domains (e.g., finance) remains underexplored. We introduce FinDeepForecast, the first live, end-to-end multi-agent system for automatically evaluating DR agents by continuously generating research-oriented financial forecasting tasks. This system is equipped with a dual-track taxonomy, enabling the dynamic generation of recurrent and non-recurrent forecasting tasks at both corporate and macro levels. With this system, we generate FinDeepForecastBench, a weekly evaluation benchmark over a ten-week horizon, encompassing 8 global economies and 1,314 listed companies, and evaluate 13 representative methods. Extensive experiments show that, while DR agents consistently outperform strong baselines, their performance still falls short of genuine forward-looking financial reasoning. We expect the proposed FinDeepForecast system to consistently facilitate future advancements of DR agents in research-oriented financial forecasting tasks. The benchmark and leaderboard are publicly available on the OpenFinArena Platform.

preprint2026arXiv

FinDeepResearch: Evaluating Deep Research Agents in Rigorous Financial Analysis

Deep Research (DR) agents, powered by advanced Large Language Models (LLMs), have recently garnered increasing attention for their capability in conducting complex research tasks. However, existing literature lacks a rigorous and systematic evaluation of DR Agent's capabilities in critical research analysis. To address this gap, we first propose HisRubric, a novel evaluation framework with a hierarchical analytical structure and a fine-grained grading rubric for rigorously assessing DR agents' capabilities in corporate financial analysis. This framework mirrors the professional analyst's workflow, progressing from data recognition to metric calculation, and finally to strategic summarization and interpretation. Built on this framework, we construct a FinDeepResearch benchmark that comprises 64 listed companies from 8 financial markets across 4 languages, encompassing a total of 15,808 grading items. We further conduct extensive experiments on the FinDeepResearch using 16 representative methods, including 6 DR agents, 5 LLMs equipped with both deep reasoning and search capabilities, and 5 LLMs with deep reasoning capabilities only. The results reveal the strengths and limitations of these approaches across diverse capabilities, financial markets, and languages, offering valuable insights for future research and development. The benchmark and evaluation code is publicly available at https://OpenFinArena.com/.

preprint2026arXiv

FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics

AI research agents accelerate ML research by automating hypothesis generation, experimentation, and empirical refinement. Existing agent strategies range from greedy hill-climbing to tree search and evolutionary optimization, yet which strategy choices drive performance remains unclear. Answering this question requires a benchmark that separates agent strategy (e.g., search topology) from execution infrastructure (e.g., code editor), so that performance differences are attributable to strategy rather than infrastructure, and that provides process-level metrics beyond final scores to analyze exploration behaviors. Existing benchmarks offer limited support. We propose FML-Bench, a benchmark of 18 fundamental ML research tasks across 10 domains that separates agent strategy from execution infrastructure and defines 12 process-level behavioral metrics. Evaluating six representative agents, we find that: (1) strategy complexity alone does not guarantee strong performance: a simple greedy hill-climber nearly matches the best-performing tree-search agent, both well above the remaining agents; (2) our analysis suggests this pattern relates to improvement opportunity structure: greedy search tends to be more effective when opportunities are dense, while tree-search and evolutionary strategies tend to be more effective when opportunities are sparse; an adaptive agent built on this insight switches to broader exploration upon detecting improvement stagnation and outperforms the other six agents, lending initial support to this observation; and (3) process-level analysis reveals that early convergence and directionally focused exploration are significantly associated with final performance, while solution diversity and compute cost are not. Our benchmark is available at: https://github.com/qrzou/FML-bench.

preprint2026arXiv

FreeText: Training-Free Text Rendering in Diffusion Transformers via Attention Localization and Spectral Glyph Injection

Large-scale text-to-image (T2I) diffusion models excel at open-domain synthesis but still struggle with precise text rendering, especially for multi-line layouts, dense typography, and long-tailed scripts such as Chinese. Prior solutions typically require costly retraining or rigid external layout constraints, which can degrade aesthetics and limit flexibility. We propose \textbf{FreeText}, a training-free, plug-and-play framework that improves text rendering by exploiting intrinsic mechanisms of \emph{Diffusion Transformer (DiT)} models. \textbf{FreeText} decomposes the problem into \emph{where to write} and \emph{what to write}. For \emph{where to write}, we localize writing regions by reading token-wise spatial attribution from endogenous image-to-text attention, using sink-like tokens as stable spatial anchors and topology-aware refinement to produce high-confidence masks. For \emph{what to write}, we introduce Spectral-Modulated Glyph Injection (SGMI), which injects a noise-aligned glyph prior with frequency-domain band-pass modulation to strengthen glyph structure and suppress semantic leakage (rendering the concept instead of the word). Extensive experiments on Qwen-Image, FLUX.1-dev, and SD3 variants across longText-Benchmark, CVTG, and our CLT-Bench show consistent gains in text readability while largely preserving semantic alignment and aesthetic quality, with modest inference overhead.

preprint2026arXiv

Generative structure search for efficient and diverse discovery of molecular and crystal structures

Predicting stable and metastable structures is central to molecular and materials discovery, but remains limited by the cost of searching high-dimensional energy landscapes. Deep generative models offer efficient structure sampling, yet their outputs remain shaped by training data and can underexplore minima that are rare but physically relevant. We introduce generative structure search (GSS), a unified framework that formulates diffusion-based generation and random structure search (RSS) as limiting regimes of a common sampling process driven by learned score fields and physical forces. Coupling these drivers lets GSS use data priors to accelerate sampling while retaining energy-guided exploration of local minima. Across molecular and crystalline systems, GSS recovers diverse metastable structures with more than tenfold lower sampling cost than RSS for broad coverage and remains effective for compositions outside the training distribution. The results establish a physically grounded generative search strategy for discovering structures beyond the reach of data-driven sampling alone.

preprint2026arXiv

GREx: Generalized Referring Expression Segmentation, Comprehension, and Generation

Referring Expression Segmentation (RES) and Comprehension (REC) respectively segment and detect the object described by an expression, while Referring Expression Generation (REG) generates an expression for the selected object. Existing datasets and methods commonly support single-target expressions only, i.e., one expression refers to one object, not considering multi-target and no-target expressions. This greatly limits the real applications of REx (RES/REC/REG). This paper introduces three new benchmarks called Generalized Referring Expression Segmentation (GRES), Comprehension (GREC), and Generation (GREG), collectively denoted as GREx, which extend the classic REx to allow expressions to identify an arbitrary number of objects. We construct the first large-scale GREx dataset gRefCOCO that contains multi-target, no-target, and single-target expressions and their corresponding images with labeled targets. GREx and gRefCOCO are designed to be backward-compatible with REx, facilitating extensive experiments to study the performance gap of the existing REx methods on GREx tasks. One of the challenges of GRES/GREC is complex relationship modeling, for which we propose a baseline ReLA that adaptively divides the image into regions with sub-instance clues and explicitly models the region-region and region-language dependencies. The proposed ReLA achieves the state-of-the-art results on the both GRES and GREC tasks. The proposed gRefCOCO dataset and method are available at https://henghuiding.github.io/GREx.

preprint2026arXiv

Improving Human Image Animation via Semantic Representation Alignment

The field of image-to-video generation has made remarkable progress. However, challenges such as human limb twisting and facial distortion persist, especially when generating long videos or modeling intensive motions. Existing human image animation works address these issues by incorporating human-specific semantic representations, e.g., dense poses or ID embeddings, as additional conditions. However, conditioning on these representations could decrease the generation flexibility. Moreover, their reliance on RGB pixel supervision also lacks emphasis on learning necessary 3D geometric relationships and temporal coherence. In contrast, we introduce a novel approach named SemanticREPA that leverages these semantic representations as supervision signals through representation alignment. Specifically, we begin by training a structure alignment module that aligns the structure representations obtained from video latents with video depth estimation features. We then fix the pretrained module, and utilize it to provide additional supervision on the structure representations of the diffusion models, achieving structure rectification to generate coherent and stable human structures. Simultaneously, we develop an ID alignment module to align the ID representations of the generated videos to face recognition features. We further propose to use the predicted structure representations to refine identity restoration in relevant regions. With structure and ID alignment, our method demonstrates superior quality on extended character motions and enhanced character consistency.

preprint2026arXiv

Learngene Search Across Multiple Datasets for Building Variable-Sized Models

Deep learning methods are widely used under diverse resource constraints, resulting in models of varying sizes, such as the Vision Transformer (ViT) series. Deploying these models typically requires costly pretraining and finetuning. The Learngene paradigm addresses this issue by extracting transferable components, called learngenes, from a pretrained ancestry model (Ans-Net) to initialize variable-sized descendant models (Des-Nets).Existing learngene extraction methods rely on a single dataset, limiting downstream performance. To address this limitation, we propose Learngene Search Across Multiple Datasets for Building Variable-Sized Models (LSAMD). LSAMD expands the Ans-Net into a searchable super Ans-Net with dataset-specific blocks and dataset adapters (DADs). During training, LSAMD searches for an optimal architecture path for each dataset. The base blocks most frequently selected across datasets are extracted as learngenes for initializing Des-Nets.Experiments on multiple datasets show that LSAMD achieves performance comparable to pretrain-finetune methods while significantly reducing storage and training costs.

preprint2026arXiv

Mamoda2.5: Enhancing Unified Multimodal Model with DiT-MoE

We present Mamoda2.5, a unified AR-Diffusion framework that seamlessly integrates multimodal understanding and generation within a single architecture. To efficiently enhance the model's generation capability, we equip the Diffusion Transformer backbone with a fine-grained Mixture-of-Experts (MoE) design (128 experts, Top-8 routing), yielding a 25B-parameter model that activates only 3B parameters, significantly reducing training costs while scaling up the model capacity. Mamoda2.5 achieves top-tier generation performance on VBench 2.0 and sets a new record in video editing quality, surpassing evaluated open-source models and matching the performance of current top-tier proprietary models, including the Kling O1 on OpenVE-Bench. Furthermore, we introduce a joint few-step distillation and reinforcement learning framework that compresses the 30-step editing model into a 4-step model and greatly accelerates model inference. Compared to open-source baselines, Mamoda2.5 achieves up to $95.9\times$ faster video editing inference. In real-world applications, Mamoda2.5 has been successfully deployed for content moderation and creative restoration tasks in advertising scenarios, achieving a 98% success rate in internal advertising video editing scenario.

preprint2026arXiv

Masked Next-Scale Prediction for Self-supervised Scene Text Recognition

Scene Text Recognition requires modeling visual structures that evolve from coarse layouts to fine-grained character strokes. Training such models relies on large amounts of annotated data. Recent self-supervised approaches, such as Masked Image Modeling (MIM), alleviate this dependency by leveraging large-scale unlabeled data. Yet most existing MIM methods operate at a single spatial scale and fail to capture the hierarchical nature of scene text. In this work, we introduce Masked Next-Scale Prediction (MNSP), a unified self-supervised framework designed to explicitly model cross-scale structural evolution. The framework incorporates Next-Scale Prediction (NSP), which learns hierarchical representations by predicting higher-resolution features from lower-resolution contexts. Naive scale prediction, however, tends to produce spatially diffuse attention, directing the model toward background regions rather than textual structures. MNSP resolves this limitation by jointly learning cross-scale prediction and masked image reconstruction. NSP captures global layout priors across resolutions, while masked reconstruction imposes strong local constraints that guide attention toward informative text regions. A Multi-scale Linguistic Alignment module further maintains semantic consistency across different resolutions. Extensive experiments demonstrate that MNSP achieves state-of-the-art performance, reaching 86.2\% average accuracy on the challenging Union14M benchmark and 96.7\% across six standard datasets. Additional analyses show that our method improves robustness under extreme scale and layout variations. Code is available at https://github.com/CzhczhcHczh/MNSP

preprint2026arXiv

Motivating Effort with Information about Future Rewards

This paper studies the optimal mechanism to motivate effort in a dynamic principal-agent model without transfers. An agent is engaged in a task with uncertain future rewards and can quit at any time. The principal knows the reward and provides information over time to motivate effort. We provide a unified framework and derive the optimal information policy in closed form across stationary and nonstationary environments. Within this framework, we identify two precise conditions, each of which guarantees that dynamic disclosure is strictly valuable. First, if the principal is impatient compared to the agent, she prefers the front-loaded effort schedule induced by dynamic disclosure; in a stationary environment, dynamic disclosure is beneficial if and only if the principal is less patient. Second, in an environment where the agent would become pessimistic over time absent any disclosure, dynamic information provision can counteract this downward trend and encourage longer effort. Notably, patience remains a crucial determinant of the structure of the optimal policy.

preprint2026arXiv

Report of the 5th PVUW Challenge: Towards More Diverse Modalities in Pixel-Level Understanding

This report summarizes the objectives, datasets, and top-performing methodologies of the 2026 Pixel-level Video Understanding in the Wild (PVUW) Challenge, hosted at CVPR 2026, which evaluates state-of-the-art models under highly unconstrained conditions. To provide a comprehensive assessment, the 2026 edition features three specialized tracks: the MOSE track for tracking objects within densely cluttered and severely occluded scenarios; the MeViS-Text track for localizing targets via motion-focused linguistic expressions; and the newly inaugurated MeViS-Audio track, which pioneers acoustic-driven object segmentation. By introducing previously unreleased challenging data and analyzing the cutting-edge, multimodal solutions submitted by participants, this report highlights the community's latest technical advancements and charts promising future directions for robust video scene comprehension.

preprint2026arXiv

SAM3-DMS: Decoupled Memory Selection for Multi-target Video Segmentation of SAM3

Segment Anything 3 (SAM3) has established a powerful foundation that robustly detects, segments, and tracks specified targets in videos. However, in its original implementation, its group-level collective memory selection is suboptimal for complex multi-object scenarios, as it employs a synchronized decision across all concurrent targets conditioned on their average performance, often overlooking individual reliability. To this end, we propose SAM3-DMS, a training-free decoupled strategy that utilizes fine-grained memory selection on individual objects. Experiments demonstrate that our approach achieves robust identity preservation and tracking stability. Notably, our advantage becomes more pronounced with increased target density, establishing a solid foundation for simultaneous multi-target video segmentation in the wild.

preprint2026arXiv

SEED: Targeted Data Selection by Weighted Independent Set

Data selection seeks to identify a compact yet informative subset from large-scale training corpora, balancing sample quality against collection diversity. We formulate this problem as a Weighted Independent Set (WIS) on a similarity graph, where nodes represent data samples weighted by influence, and edges connect semantically redundant pairs. This formulation naturally yields subsets that are simultaneously high-quality and diverse. However, two challenges arise in practice: naive node weights fail to distinguish informative signals from gradient noise, and edge construction under heterogeneous domain distributions produces structurally imbalanced graphs that bias selection toward sparse regions. To address these issues, we introduce two principled refinements from a unified graph perspective: (1) \textit{node value calibration} that restricts influence estimation to the bilateral salient subspace to ground node importance in task-relevant signals rather than surface-level statistics; (2) \textit{local scale normalization} that adapts edge thresholds to local neighborhood density, mitigating graph imbalance induced by cross-domain distribution shifts. Together, these components yield a robust and scalable data selection pipeline dubbed SEED. We further construct \texttt{Honeybee-Remake-SEED-200K}, a compact multimodal dataset curated by SEED. Extensive experiments show that SEED consistently outperforms state-of-the-art methods on instruction tuning, visual instruction tuning, and semantic segmentation across diverse model families.

preprint2026arXiv

Self-organized MT Direction Maps Emerge from Spatiotemporal Contrastive Optimization

The spatial and functional organization of the primate visual cortex is a fundamental problem in neuroscience. While recent computational frameworks like the Topographic Deep Artificial Neural Network (TDANN) have successfully modeled spatial organization in the ventral stream, the computational origins of the dorsal stream's distinct topographies, such as direction-selective maps in the middle temporal (MT) area, remain largely unresolved. In this work, we present a spatiotemporal TDANN to investigate whether MT topography is governed by the same universal principles. By training a 3D ResNet on naturalistic videos via a Momentum Contrast (MoCo) self-supervised paradigm alongside a biologically inspired spatial loss, we demonstrate the spontaneous emergence of brain-like direction maps and topological pinwheel structures. Crucially, we reveal that MT tuning properties, characterized by strong direction selectivity paired with a residual axial component, arise from a strict optimization trade-off between task-driven discriminative pressure and spatial regularization. The model's representations quantitatively match in vivo macaque MT physiological baselines, including direction selectivity index, circular variance, and pinwheel density. These findings unify the computational origins of the ventral and dorsal streams, establishing a general mechanism for cortical self-organization.

preprint2026arXiv

Structure preservation and emergent dissipation in stochastic wave equations with transport noise

We study nonlinear wave equations perturbed by transport noise acting either on the displacement or on the velocity. Such noise models random advection and, under suitable scaling of space covariance, may generate an effective dissipative term. We establish well-posedness in both cases and analyse the associated scaling limits. When the noise acts on the displacement, the system preserves its original structure and converges to the deterministic nonlinear wave equation, whereas if it acts on the velocity, the rescaled dynamics produce an additional Laplacian damping term, leading to a stochastic derivation of a Westervelt-type acoustic model.

preprint2026arXiv

Understanding Cross-Language Transfer Improvements in Low-Resource HTR: The Role of Sequence Modeling

Handwritten Text Recognition (HTR) for Arabic-script languages benefits from cross-language joint training under low-resource conditions, particularly when using CRNN-based models that combine convolutional encoders with sequence modeling. However, it remains unclear whether these improvements are better explained by shared visual representations or sequence-level dependencies. In this work, we conduct a controlled architectural study of line-level Arabic-script HTR, comparing CNN-only models with CTC decoding and CRNN models under identical single-script and multi-script training regimes. Experiments are performed on Arabic (KHATT), Urdu (NUST-UHWR), and Persian (PHTD) datasets under low-resource settings (K in {100, 500, 1000}). Our results show a clear divergence in transfer behavior: while CNN-only models exhibit limited or unstable improvements, CRNN models achieve better performance under multi-script training, particularly in the most data-constrained regimes. Focusing on transfer improvements (delta CER) rather than absolute performance, we find that cross-language improvements are associated with sequence-level modeling, while sharing visual representations learned by the CNN encoder, corresponding to similarities in character shapes across scripts, alone appears to be insufficient. This finding suggests that contextual modeling plays an important role in enabling effective transfer in low-resource scenarios, and that similar behavior may extend to other low-resource language settings.

preprint2026arXiv

Understanding Performance Collapse in Layer-Pruned Large Language Models via Decision Representation Transitions

Layer pruning efficiently reduces Large Language Model (LLM) computational costs but often triggers sudden performance collapse. Existing representation-based analyses struggle to explain this mechanism. We propose studying pruning through decision representation. Focusing on multiple-choice tasks, we introduce two metrics, Decision Margin and Option Frequency, and an Iterative Pruning method to analyze layer-wise decision dynamics. Our findings reveal a sharp decision transition that partitions the network into two stages: a Silent Phase, where the model cannot yet predict the correct answer, and a Decisive Phase, where the correct prediction emerges. We also find that pruning the Decisive Phase has minimal impact, whereas pruning the Silent Phase triggers immediate performance collapse, highlighting its extreme sensitivity to structural changes. Therefore, we conclude that pruning-induced collapse stems from disrupting the Silent Phase, which prevents the critical decision transition from occurring.

preprint2026arXiv

Video Generation with Predictive Latents

Video Variational Autoencoder (VAE) enables latent video generative modeling by mapping the visual world into compact spatiotemporal latent spaces, improving training efficiency and stability. While existing video VAEs achieve commendable reconstruction quality, continued optimization of reconstruction does not necessarily translate into improved generative performance. How to enhance the diffusability of video latents remains a critical and unresolved challenge. In this work, inspired by principles of predictive world modeling, we investigate the potential of predictive learning to improve the video generative modeling. To this end, we introduce a simple and effective predictive reconstruction objective that unifies predictive learning with video reconstruction. Specifically, we randomly discard future frames and encode only partial past observations, while training the decoder to reconstruct the observed frames and predict future ones simultaneously. This design encourages the latent space to encode temporally predictive structures and build a more coherent understanding of video dynamics, thereby improving generation quality. Our model, termed Predictive Video VAE (PV-VAE), achieves superior performance on video generation, with 52% faster convergence and a 34.42 FVD improvement over the Wan2.2 VAE on UCF101. Furthermore, comprehensive analyses demonstrate that PV-VAE not only exhibits favorable scalability, with generative performance improving alongside VAE training, but also yields consistent gains in downstream video understanding, underscoring a latent space that effectively captures temporal coherence and motion priors.

preprint2026arXiv

ViewSAM: Learning View-aware Cross-modal Semantics for Weakly Supervised Cross-view Referring Multi-Object Tracking

Cross-view Referring Multi-Object Tracking (CRMOT) aims to track multiple objects specified by natural language across multiple camera views, with globally consistent identities. Despite recent progress, existing methods rely heavily on costly frame-level spatial annotations and cross-view identity supervision. To reduce such reliance, we explore CRMOT under weak supervision by leveraging the capabilities of foundation models. However, our empirical study shows that directly applying foundation models such as SAM2 and SAM3, even with task-specific modifications, fails to accurately understand referring expressions and maintain consistent identities across views. Yet, they remain effective at producing reliable object tracklets that can serve as pseudo supervision. We therefore repurpose foundation models as pseudo-label generators and propose a two-stage framework for weakly supervised CRMOT, using only object category labels as coarse-grained supervision. In the first stage, we design an Affinity-guided Cross-view Re-prompting strategy to refine and associate SAM3-generated tracklets across cameras, producing reliable cross-view pseudo labels for subsequent training. In the second stage, we introduce ViewSAM, a CRMOT model built upon SAM2 that explicitly models view-aware cross-modal semantics. By formulating view-induced variations as learnable conditions, ViewSAM bridges the gap between view-variant visual observations and view-invariant textual expressions, enabling robust cross-view referring tracking with only approximately 10% additional parameters. Extensive experiments demonstrate that ViewSAM achieves SOTA performance under weak supervision and remains competitive with fully supervised methods.

preprint2026arXiv

VLingNav: Embodied Navigation with Adaptive Reasoning and Visual-Assisted Linguistic Memory

VLA models have shown promising potential in embodied navigation by unifying perception and planning while inheriting the strong generalization abilities of large VLMs. However, most existing VLA models rely on reactive mappings directly from observations to actions, lacking the explicit reasoning capabilities and persistent memory required for complex, long-horizon navigation tasks. To address these challenges, we propose VLingNav, a VLA model for embodied navigation grounded in linguistic-driven cognition. First, inspired by the dual-process theory of human cognition, we introduce an adaptive chain-of-thought mechanism, which dynamically triggers explicit reasoning only when necessary, enabling the agent to fluidly switch between fast, intuitive execution and slow, deliberate planning. Second, to handle long-horizon spatial dependencies, we develop a visual-assisted linguistic memory module that constructs a persistent, cross-modal semantic memory, enabling the agent to recall past observations to prevent repetitive exploration and infer movement trends for dynamic environments. For the training recipe, we construct Nav-AdaCoT-2.9M, the largest embodied navigation dataset with reasoning annotations to date, enriched with adaptive CoT annotations that induce a reasoning paradigm capable of adjusting both when to think and what to think about. Moreover, we incorporate an online expert-guided reinforcement learning stage, enabling the model to surpass pure imitation learning and to acquire more robust, self-explored navigation behaviors. Extensive experiments demonstrate that VLingNav achieves state-of-the-art performance across a wide range of embodied navigation benchmarks. Notably, VLingNav transfers to real-world robotic platforms in a zero-shot manner, executing various navigation tasks and demonstrating strong cross-domain and cross-task generalization.

preprint2026arXiv

WaveFormer: Frequency-Time Decoupled Vision Modeling with Wave Equation

Vision modeling has advanced rapidly with Transformers, whose attention mechanisms capture visual dependencies but lack a principled account of how semantic information propagates spatially. We revisit this problem from a wave-based perspective: feature maps are treated as spatial signals whose evolution over an internal propagation time (aligned with network depth) is governed by an underdamped wave equation. In this formulation, spatial frequency-from low-frequency global layout to high-frequency edges and textures-is modeled explicitly, and its interaction with propagation time is controlled rather than implicitly fixed. We derive a closed-form, frequency-time decoupled solution and implement it as the Wave Propagation Operator (WPO), a lightweight module that models global interactions in O(N log N) time-far lower than attention. Building on WPO, we propose a family of WaveFormer models as drop-in replacements for standard ViTs and CNNs, achieving competitive accuracy across image classification, object detection, and semantic segmentation, while delivering up to 1.6x higher throughput and 30% fewer FLOPs than attention-based alternatives. Furthermore, our results demonstrate that wave propagation introduces a complementary modeling bias to heat-based methods, effectively capturing both global coherence and high-frequency details essential for rich visual semantics. Codes are available at: https://github.com/ZishanShu/WaveFormer.

preprint2026arXiv

XPERT: Expert Knowledge Transfer for Effective Training of Language Models

Mixture-of-Experts (MoE) language models organize knowledge into explicitly routed expert modules, making expert-level representations traceable and analyzable. By analyzing expert activation patterns in MoE large language models (LLMs), we find that a subset of experts is consistently activated across diverse knowledge domains. These common experts encode cross-domain, generalizable knowledge that is closely related to model generalization, naturally raising the question of how such identifiable expert knowledge can be practically reused. Motivated by this observation, we propose XPERT, a framework that extracts, consolidates, and reuses expert knowledge from pre-trained MoE LLMs to support more effective training of language models across different model scales. XPERT identifies cross-domain experts via inference-only analysis, refines their representations through tensor decomposition, and adapts the extracted knowledge to reuse in downstream models. Experiments on language understanding and dialogue generation benchmarks show that models benefiting from reused expert knowledge achieve consistently stronger performance and faster convergence compared to strong baselines. These results highlight MoE LLMs as structured and reusable knowledge sources, and demonstrate the value of expert-level knowledge reuse for improving model training.

preprint2025arXiv

AINav: Large Language Model-Based Adaptive Interactive Navigation

Robotic navigation in complex environments remains a critical research challenge. Traditional navigation methods focus on optimal trajectory generation within fixed free workspace, therefore struggling in environments lacking viable paths to the goal, such as disaster zones or cluttered warehouses. To address this problem, we propose AINav, an adaptive interactive navigation approach that proactively interacts with environments to create feasible paths to achieve originally unreachable goals. Specifically, we present a primitive skill tree for task planning with large language models (LLMs), facilitating effective reasoning to determine interaction objects and sequences. To ensure robust subtask execution, we adopt reinforcement learning to pre-train a comprehensive skill library containing versatile locomotion and interaction behaviors for motion planning. Furthermore, we introduce an adaptive replanning approach featuring two LLM-based modules: an advisor serving as a flexible replanning trigger and an arborist for autonomous plan adjustment. Integrated with the tree structure, the replanning mechanism allows for convenient node addition and pruning, enabling rapid plan adaptation in a priori unknown environments. Comprehensive simulations and experiments have demonstrated AINav's effectiveness and adaptivity in diverse scenarios. The supplementary video is available at: https://youtu.be/CjXm5KFx9AI.

preprint2025arXiv

Influence of ambient temperature on cavitation bubble dynamics

We investigate the influence of ambient temperature on the dynamics of spark-generated cavitation bubbles over a broad temperature range of 23 to 90$^\circ \text{C}$. Increasing temperature, the attenuation of collapse intensity of a bubble in a free field is quantitatively characterised through the Rayleigh factor, minimum bubble volume, and maximum collapse velocity. In scenarios where the bubble is initiated near a rigid boundary, this temperature-dependent weakening effect manifests further as a reduction in jet velocity and bubble migration. Additionally, our findings demonstrate that when ambient temperature exceeds 70$^\circ \text{C}$, secondary cavitation forms near the bubble surface around the moment of maximum bubble expansion, followed by coalescence-induced surface wrinkles. These perturbations trigger Rayleigh-Taylor instability and enhance bubble fission. We determine the internal gas pressure of the bubble at its maximum expansion via the Rayleigh-Plesset equation with the input of bubble radius from experimental measurements. It reveals that the secondary cavitation is derived from the gas pressure descending below the saturated vapor pressure, which provides nucleation-favorable conditions. This study sheds light on the physics behind erosion mitigation in high-temperature fluids from the perspective of cavitation bubble dynamics.

preprint2025arXiv

MiMo-Audio: Audio Language Models are Few-Shot Learners

Existing audio language models typically rely on task-specific fine-tuning to accomplish particular audio tasks. In contrast, humans are able to generalize to new audio tasks with only a few examples or simple instructions. GPT-3 has shown that scaling next-token prediction pretraining enables strong generalization capabilities in text, and we believe this paradigm is equally applicable to the audio domain. By scaling MiMo-Audio's pretraining data to over one hundred million of hours, we observe the emergence of few-shot learning capabilities across a diverse set of audio tasks. We develop a systematic evaluation of these capabilities and find that MiMo-Audio-7B-Base achieves SOTA performance on both speech intelligence and audio understanding benchmarks among open-source models. Beyond standard metrics, MiMo-Audio-7B-Base generalizes to tasks absent from its training data, such as voice conversion, style transfer, and speech editing. MiMo-Audio-7B-Base also demonstrates powerful speech continuation capabilities, capable of generating highly realistic talk shows, recitations, livestreaming and debates. At the post-training stage, we curate a diverse instruction-tuning corpus and introduce thinking mechanisms into both audio understanding and generation. MiMo-Audio-7B-Instruct achieves open-source SOTA on audio understanding benchmarks (MMSU, MMAU, MMAR, MMAU-Pro), spoken dialogue benchmarks (Big Bench Audio, MultiChallenge Audio) and instruct-TTS evaluations, approaching or surpassing closed-source models. Model checkpoints and full evaluation suite are available at https://github.com/XiaomiMiMo/MiMo-Audio.

preprint2025arXiv

MTSP-LDP: A Framework for Multi-Task Streaming Data Publication under Local Differential Privacy

The proliferation of streaming data analytics in data-driven applications raises critical privacy concerns, as directly collecting user data may compromise personal privacy. Although existing $w$-event local differential privacy (LDP) mechanisms provide formal guarantees without relying on trusted third parties, their practical deployment is hindered by two key limitations. First, these methods are designed primarily for publishing simple statistics at each timestamp, making them inherently unsuitable for complex queries. Second, they handle data at each timestamp independently, failing to capture temporal correlations and consequently degrading the overall utility. To address these issues, we propose MTSP-LDP, a novel framework for \textbf{M}ulti-\textbf{T}ask \textbf{S}treaming data \textbf{P}ublication under $w$-event LDP. MTSP-LDP adopts an \emph{Optimal Privacy Budget Allocation} algorithm to dynamically allocate privacy budgets by analyzing temporal correlations within each window. It then constructs a \emph{data-adaptive private binary tree structure} to support complex queries, which is further refined by cross-timestamp grouping and smoothing operations to enhance estimation accuracy. Furthermore, a unified \emph{Budget-Free Multi-Task Processing} mechanism is introduced to support a variety of streaming queries without consuming additional privacy budget. Extensive experiments on real-world datasets demonstrate that MTSP-LDP consistently achieves high utility across various streaming tasks, significantly outperforming existing methods.

preprint2025arXiv

ReSPIRe: Informative and Reusable Belief Tree Search for Robot Probabilistic Search and Tracking in Unknown Environments

Target search and tracking (SAT) is a fundamental problem for various robotic applications such as search and rescue and environmental exploration. This paper proposes an informative trajectory planning approach, namely ReSPIRe, for SAT in unknown cluttered environments under considerably inaccurate prior target information and limited sensing field of view. We first develop a novel sigma point-based approximation approach to fast and accurately estimate mutual information reward under non-Gaussian belief distributions, utilizing informative sampling in state and observation spaces to mitigate the computational intractability of integral calculation. To tackle significant uncertainty associated with inadequate prior target information, we propose the hierarchical particle structure in ReSPIRe, which not only extracts critical particles for global route guidance, but also adjusts the particle number adaptively for planning efficiency. Building upon the hierarchical structure, we develop the reusable belief tree search approach to build a policy tree for online trajectory planning under uncertainty, which reuses rollout evaluation to improve planning efficiency. Extensive simulations and real-world experiments demonstrate that ReSPIRe outperforms representative benchmark methods with smaller MI approximation error, higher search efficiency, and more stable tracking performance, while maintaining outstanding computational efficiency.

preprint2025arXiv

Tailoring Neel orders in Layered Topological Antiferromagnets

In the two-dimensional limit, the interplay between Neel order and band topology in van der Waals topological antiferromagnets can give rise to novel quantum phenomena in the quantum anomalous Hall state, including the cascaded quantum phase transition and spin-modulation effect. However, due to the absence of net magnetization in antiferromagnets, probing the energetically degenerate Neel orders has long remained a significant challenge. Inspired by recent advances in realizing the quantum anomalous Hall effect in AlOx-capped layered topological antiferromagnet MnBi2Te4, we demonstrate deterministic control over the Neel order through surface anisotropy engineering enabled by the AlOx capping layer. By tuning the surface anisotropy, we uncover paritydependent symmetry breaking states that manifest as distinct odd-even boundary architectures, including 180 degree domain walls or continuous spin structures. Comparative studies between AlOx-capped and pristine odd-layer MnBi2Te4 flakes using domain-resolved magnetic force microscopy reveal pronounced differences in coercivity and magnetization-reversal dynamics. Notably, an unconventional giant exchange bias, which arises from perpendicular magnetic anisotropy rather than traditional interface pinning mechanisms, is observed for the first time. Our findings establish a pathway for manipulating Neel order through surface modification in A-type antiferromagnets, offering new opportunities for spintronic devices and quantum information technologies.

preprint2024arXiv

Foundation Model for Endoscopy Video Analysis via Large-scale Self-supervised Pre-train

Foundation models have exhibited remarkable success in various applications, such as disease diagnosis and text report generation. To date, a foundation model for endoscopic video analysis is still lacking. In this paper, we propose Endo-FM, a foundation model specifically developed using massive endoscopic video data. First, we build a video transformer, which captures both local and global long-range dependencies across spatial and temporal dimensions. Second, we pre-train our transformer model using global and local views via a self-supervised manner, aiming to make it robust to spatial-temporal variations and discriminative across different scenes. To develop the foundation model, we construct a large-scale endoscopy video dataset by combining 9 publicly available datasets and a privately collected dataset from Baoshan Branch of Renji Hospital in Shanghai, China. Our dataset overall consists of over 33K video clips with up to 5 million frames, encompassing various protocols, target organs, and disease types. Our pre-trained Endo-FM can be easily adopted for a given downstream task via fine-tuning by serving as the backbone. With experiments on 3 different types of downstream tasks, including classification, segmentation, and detection, our Endo-FM surpasses the current state-of-the-art (SOTA) self-supervised pre-training and adapter-based transfer learning methods by a significant margin, such as VCL (3.1% F1, 4.8% Dice, and 5.5% F1 for classification, segmentation, and detection) and ST-Adapter (5.9% F1, 9.6% Dice, and 9.9% F1 for classification, segmentation, and detection). Code, datasets, and models are released at https://github.com/med-air/Endo-FM.

preprint2024arXiv

LaCon: Late-Constraint Diffusion for Steerable Guided Image Synthesis

Diffusion models have demonstrated impressive abilities in generating photo-realistic and creative images. To offer more controllability for the generation process, existing studies, termed as early-constraint methods in this paper, leverage extra conditions and incorporate them into pre-trained diffusion models. Particularly, some of them adopt condition-specific modules to handle conditions separately, where they struggle to generalize across other conditions. Although follow-up studies present unified solutions to solve the generalization problem, they also require extra resources to implement, e.g., additional inputs or parameter optimization, where more flexible and efficient solutions are expected to perform steerable guided image synthesis. In this paper, we present an alternative paradigm, namely Late-Constraint Diffusion (LaCon), to simultaneously integrate various conditions into pre-trained diffusion models. Specifically, LaCon establishes an alignment between the external condition and the internal features of diffusion models, and utilizes the alignment to incorporate the target condition, guiding the sampling process to produce tailored results. Experimental results on COCO dataset illustrate the effectiveness and superior generalization capability of LaCon under various conditions and settings. Ablation studies investigate the functionalities of different components in LaCon, and illustrate its great potential to serve as an efficient solution to offer flexible controllability for diffusion models.

preprint2023arXiv

Benchmarking Graphormer on Large-Scale Molecular Modeling Datasets

This technical note describes the recent updates of Graphormer, including architecture design modifications, and the adaption to 3D molecular dynamics simulation. With these simple modifications, Graphormer could attain better results on large-scale molecular modeling datasets than the vanilla one, and the performance gain could be consistently obtained on 2D and 3D molecular graph modeling tasks. In addition, we show that with a global receptive field and an adaptive aggregation strategy, Graphormer is more powerful than classic message-passing-based GNNs. Empirically, Graphormer could achieve much less MAE than the originally reported results on the PCQM4M quantum chemistry dataset used in KDD Cup 2021. In the meanwhile, it greatly outperforms the competitors in the recent Open Catalyst Challenge, which is a competition track on NeurIPS 2021 workshop, and aims to model the catalyst-adsorbate reaction system with advanced AI models. All codes could be found at https://github.com/Microsoft/Graphormer.

preprint2023arXiv

Unified Gas-Kinetic Particle Method for Frequency-dependent Radiation Transport

This paper proposes a unified gas-kinetic particle (UGKP) method for the frequency-dependent photon transport process. The photon transport is a typical multiscale process governed by the nonlinear radiative transfer equations (RTE). The flow regime of photon transport varies from the ballistic regime to the diffusive regime with respect to optical depth and photon frequency. The UGKP method is an asymptotic preserving (AP) scheme and a regime adaptive scheme. The teleportation error is significantly reduced, and the computational efficiency is remarkably improved in the diffusive regime. Distinguished from the standard multigroup treatment, the proposed UGKP method solves the frequency space in a non-discretized way. Therefore the Rosseland diffusion system can be precisely preserved in the optically thick regime. Taking advantage of the local integral solution of RTE, the distribution of the emitted photon can be constructed from its macroscopic moments. The Monte Carlo particles in the UGKP method need only be tracked before their first collision events, and a re-sampling process is performed to close the photon distribution for each time step. The large computational cost of excessive scattering events can be saved, especially in the optically thick regime. The proposed UGKP method is implicit and removes the light speed constraint on the time step. The particle tracking approach combining the implicit formulation makes the proposed UGKP method an efficient solution algorithm for frequency-dependent radiative transfer problems. We demonstrate with numerical examples the capability of the proposed multi-frequency UGKP method.

preprint2022arXiv

6GAN: IPv6 Multi-Pattern Target Generation via Generative Adversarial Nets with Reinforcement Learning

Global IPv6 scanning has always been a challenge for researchers because of the limited network speed and computational power. Target generation algorithms are recently proposed to overcome the problem for Internet assessments by predicting a candidate set to scan. However, IPv6 custom address configuration emerges diverse addressing patterns discouraging algorithmic inference. Widespread IPv6 alias could also mislead the algorithm to discover aliased regions rather than valid host targets. In this paper, we introduce 6GAN, a novel architecture built with Generative Adversarial Net (GAN) and reinforcement learning for multi-pattern target generation. 6GAN forces multiple generators to train with a multi-class discriminator and an alias detector to generate non-aliased active targets with different addressing pattern types. The rewards from the discriminator and the alias detector help supervise the address sequence decision-making process. After adversarial training, 6GAN's generators could keep a strong imitating ability for each pattern and 6GAN's discriminator obtains outstanding pattern discrimination ability with a 0.966 accuracy. Experiments indicate that our work outperformed the state-of-the-art target generation algorithms by reaching a higher-quality candidate set.

preprint2022arXiv

A Comprehensive Study of Accelerating IPv6 Deployment

Since the lack of IPv6 network development, China is currently accelerating IPv6 deployment. In this scenario, traffic and network structure show a huge shift. However, due to the long-term prosperity, we are ignorant of the problems behind such outbreak of traffic and performance improvement events in accelerating deployment. IPv6 development in some regions will still face similar challenges in the future. To contribute to solving this problem, in this paper, we produce a new measurement framework and implement a 5-month passive measurement on the IPv6 network during the accelerating deployment in China. We combine 6 global-scale datasets to form the normal status of IPv6 network, which is against to the accelerating status formed by the passive traffic. Moreover, we compare with the traffic during World IPv6 Day 2011 and Launch 2012 to discuss the common nature of accelerating deployment. Finally, the results indicate that the IPv6 accelerating deployment is often accompanied by an unbalanced network status. It exposes unresolved security issues including the challenge of user privacy and inappropriate access methods. According to the investigation, we point the future IPv6 development after accelerating deployment.

preprint2022arXiv

A Survey for Solving Mixed Integer Programming via Machine Learning

This paper surveys the trend of leveraging machine learning to solve mixed integer programming (MIP) problems. Theoretically, MIP is an NP-hard problem, and most of the combinatorial optimization (CO) problems can be formulated as the MIP. Like other CO problems, the human-designed heuristic algorithms for MIP rely on good initial solutions and cost a lot of computational resources. Therefore, we consider applying machine learning methods to solve MIP, since ML-enhanced approaches can provide the solution based on the typical patterns from the historical data. In this paper, we first introduce the formulation and preliminaries of MIP and several traditional algorithms to solve MIP. Then, we advocate further promoting the different integration of machine learning and MIP and introducing related learning-based methods, which can be classified into exact algorithms and heuristic algorithms. Finally, we propose the outlook for learning-based MIP solvers, direction towards more combinatorial optimization problems beyond MIP, and also the mutual embrace of traditional solvers and machine learning components.

preprint2022arXiv

An efficient and easy-to-extend Matlab code of the Moving Morphable Component (MMC) method for three-dimensional topology optimization

Explicit topology optimization methods have received ever-increasing interest in recent years. In particular, a 188-line Matlab code of the two-dimensional (2D) Moving Morphable Component (MMC)-based topology optimization method was released by Zhang et al. (Struct Multidiscip Optim 53(6):1243-1260, 2016). The present work aims to propose an efficient and easy-to-extend 256-line Matlab code of the MMC method for three-dimensional (3D) topology optimization implementing some new numerical techniques. To be specific, by virtue of the function aggregation technique, accurate sensitivity analysis, which is also easy-to-extend to other problems, is achieved. Besides, based on an efficient identification algorithm for load transmission path, the degrees of freedoms (DOFs) not belonging to the load transmission path are removed in finite element analysis (FEA), which significantly accelerates the optimization process. As a result, compared to the corresponding 188-line 2D code, the performance of the optimization results, the computational efficiency of FEA, and the convergence rate and the robustness of optimization process are greatly improved. For the sake of completeness, a refined 218-line Matlab code implementing the 2D-MMC method is also provided.

preprint2022arXiv

An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

This technical note describes the recent updates of Graphormer, including architecture design modifications, and the adaption to 3D molecular dynamics simulation. The "Graphormer-V2" could attain better results on large-scale molecular modeling datasets than the vanilla one, and the performance gain could be consistently obtained on downstream tasks. In addition, we show that with a global receptive field and an adaptive aggregation strategy, Graphormer is more powerful than classic message-passing-based GNNs. Graphormer-V2 achieves much less MAE than the vanilla Graphormer on the PCQM4M quantum chemistry dataset used in KDD Cup 2021, where the latter one won the first place in this competition. In the meanwhile, Graphormer-V2 greatly outperforms the competitors in the recent Open Catalyst Challenge, which is a competition track on NeurIPS 2021 workshop, and aims to model the catalyst-adsorbate reaction system with advanced AI models. All models could be found at \url{https://github.com/Microsoft/Graphormer}.

preprint2022arXiv

An Extendable Maneuver Management Framework with Fault-Tolerant Mechanism for Vehicle Platoon Control System in Highway Scenario

Vehicle platoon often face the problem of lack of scalability of maneuvers in practical applications. Once a new scenario is added, the original program may no longer be available. To deal with this problem, this paper introduces a two-dimensional maneuver management framework with a fault-tolerant mechanism on the basis of the proposed hierarchical architecture for the platoon control system. Maneuvers and roles are two dimensions, based on which the management strategies are decoupled. This makes each vehicle in the platoon has the ability to execute management strategies of various maneuvers and the new maneuver could be extended without revising the existing part. The fault-tolerant mechanism is designed as a maneuver triggered by hardware failures to keep safe before taking over. Furthermore, three typical maneuvers are selected for case studies to illustrate how the management strategies in this framework work. Finally, a comprehensive simulation scenario integrating different maneuvers is designed and a real-world implementation using micro-vehicles is conducted. Results show that the propose two-dimensional framework could effectively deal with various maneuvers and satisfy the computational real-time requirements

preprint2022arXiv

Atom-Specific Probing of Electron Dynamics in an Atomic Adsorbate by Time-Resolved X-ray Spectroscopy

The electronic excitation occurring on adsorbates at ultrafast time scales from optical lasers that initiate surface chemical reactions is still an open question. Here, we report the ultrafast temporal evolution of X-ray absorption spectroscopy (XAS) and X-ray emission spectroscopy (XES) of a simple well known adsorbate prototype system, namely carbon (C) atoms adsorbed on a nickel (Ni(100)) surface, following intense laser optical pumping at 400 nm. We observe ultrafast (~100 fs) changes in both XAS and XES showing clear signatures of the formation of a hot electron-hole pair distribution on the adsorbate. This is followed by slower changes on a few ps time scale, shown to be consistent with thermalization of the complete C/Ni system. Density functional theory spectrum simulations support this interpretation.

preprint2022arXiv

Bayesian Sequential Stacking Algorithm for Concurrently Designing Molecules and Synthetic Reaction Networks

In the last few years, de novo molecular design using machine learning has made great technical progress but its practical deployment has not been as successful. This is mostly owing to the cost and technical difficulty of synthesizing such computationally designed molecules. To overcome such barriers, various methods for synthetic route design using deep neural networks have been studied intensively in recent years. However, little progress has been made in designing molecules and their synthetic routes simultaneously. Here, we formulate the problem of simultaneously designing molecules with the desired set of properties and their synthetic routes within the framework of Bayesian inference. The design variables consist of a set of reactants in a reaction network and its network topology. The design space is extremely large because it consists of all combinations of purchasable reactants, often in the order of millions or more. In addition, the designed reaction networks can adopt any topology beyond simple multistep linear reaction routes. To solve this hard combinatorial problem, we present a powerful sequential Monte Carlo algorithm that recursively designs a synthetic reaction network by sequentially building up single-step reactions. In a case study of designing drug-like molecules based on commercially available compounds, compared with heuristic combinatorial search methods, the proposed method shows overwhelming performance in terms of computational efficiency and coverage and novelty with respect to existing compounds.

preprint2022arXiv

Constrained Deep One-Class Feature Learning For Classifying Imbalanced Medical Images

Medical image data are usually imbalanced across different classes. One-class classification has attracted increasing attention to address the data imbalance problem by distinguishing the samples of the minority class from the majority class. Previous methods generally aim to either learn a new feature space to map training samples together or to fit training samples by autoencoder-like models. These methods mainly focus on capturing either compact or descriptive features, where the information of the samples of a given one class is not sufficiently utilized. In this paper, we propose a novel deep learning-based method to learn compact features by adding constraints on the bottleneck features, and to preserve descriptive features by training an autoencoder at the same time. Through jointly optimizing the constraining loss and the autoencoder's reconstruction loss, our method can learn more relevant features associated with the given class, making the majority and minority samples more distinguishable. Experimental results on three clinical datasets (including the MRI breast images, FFDM breast images and chest X-ray images) obtains state-of-art performance compared to previous methods.

preprint2022arXiv

Crystal structure prediction with machine learning-based element substitution

The prediction of energetically stable crystal structures formed by a given chemical composition is a central problem in solid-state physics. In principle, the crystalline state of assembled atoms can be determined by optimizing the energy surface, which in turn can be evaluated using first-principles calculations. However, performing the iterative gradient descent on the potential energy surface using first-principles calculations is prohibitively expensive for complex systems, such as those with many atoms per unit cell. Here, we present a unique methodology for crystal structure prediction (CSP) that relies on a machine learning algorithm called metric learning. It is shown that a binary classifier, trained on a large number of already identified crystal structures, can determine the isomorphism of crystal structures formed by two given chemical compositions with an accuracy of approximately 96.4\%. For a given query composition with an unknown crystal structure, the model is used to automatically select from a crystal structure database a set of template crystals with nearly identical stable structures to which element substitution is to be applied. Apart from the local relaxation calculation of the identified templates, the proposed method does not use ab initio calculations. The potential of this substation-based CSP is demonstrated for a wide variety of crystal systems.

preprint2022arXiv

Direct Molecular Conformation Generation

Molecular conformation generation aims to generate three-dimensional coordinates of all the atoms in a molecule and is an important task in bioinformatics and pharmacology. Previous methods usually first predict the interatomic distances, the gradients of interatomic distances or the local structures (e.g., torsion angles) of a molecule, and then reconstruct its 3D conformation. How to directly generate the conformation without the above intermediate values is not fully explored. In this work, we propose a method that directly predicts the coordinates of atoms: (1) the loss function is invariant to roto-translation of coordinates and permutation of symmetric atoms; (2) the newly proposed model adaptively aggregates the bond and atom information and iteratively refines the coordinates of the generated conformation. Our method achieves the best results on GEOM-QM9 and GEOM-Drugs datasets. Further analysis shows that our generated conformations have closer properties (e.g., HOMO-LUMO gap) with the groundtruth conformations. In addition, our method improves molecular docking by providing better initial conformations. All the results demonstrate the effectiveness of our method and the great potential of the direct approach. The code is released at https://github.com/DirectMolecularConfGen/DMCG

preprint2022arXiv

Disjoint Masking with Joint Distillation for Efficient Masked Image Modeling

Masked image modeling (MIM) has shown great promise for self-supervised learning (SSL) yet been criticized for learning inefficiency. We believe the insufficient utilization of training signals should be responsible. To alleviate this issue, we introduce a conceptually simple yet learning-efficient MIM training scheme, termed Disjoint Masking with Joint Distillation (DMJD). For disjoint masking (DM), we sequentially sample multiple masked views per image in a mini-batch with the disjoint regulation to raise the usage of tokens for reconstruction in each image while keeping the masking rate of each view. For joint distillation (JD), we adopt a dual branch architecture to respectively predict invisible (masked) and visible (unmasked) tokens with superior learning targets. Rooting in orthogonal perspectives for training efficiency improvement, DM and JD cooperatively accelerate the training convergence yet not sacrificing the model generalization ability. Concretely, DM can train ViT with half of the effective training epochs (3.7 times less time-consuming) to report competitive performance. With JD, our DMJD clearly improves the linear probing classification accuracy over ConvMAE by 5.8%. On fine-grained downstream tasks like semantic segmentation, object detection, etc., our DMJD also presents superior generalization compared with state-of-the-art SSL methods. The code and model will be made public at https://github.com/mx-mark/DMJD.

preprint2022arXiv

DMF-Net: Dual-Branch Multi-Scale Feature Fusion Network for copy forgery identification of anti-counterfeiting QR code

Anti-counterfeiting QR codes are widely used in people's work and life, especially in product packaging. However, the anti-counterfeiting QR code has the risk of being copied and forged in the circulation process. In reality, copying is usually based on genuine anti-counterfeiting QR codes, but the brands and models of copiers are diverse, and it is extremely difficult to determine which individual copier the forged anti-counterfeiting code come from. In response to the above problems, this paper proposes a method for copy forgery identification of anti-counterfeiting QR code based on deep learning. We first analyze the production principle of anti-counterfeiting QR code, and convert the identification of copy forgery to device category forensics, and then a Dual-Branch Multi-Scale Feature Fusion network is proposed. During the design of the network, we conducted a detailed analysis of the data preprocessing layer, single-branch design, etc., combined with experiments, the specific structure of the dual-branch multi-scale feature fusion network is determined. The experimental results show that the proposed method has achieved a high accuracy of copy forgery identification, which exceeds the current series of methods in the field of image forensics.

preprint2022arXiv

Electromagnetic Follow-up Observations of Binary Neutron Star Mergers with Early Warnings from Decihertz Gravitational-wave Observatories

We investigate the prospects of electromagnetic follow-up observations for binary neutron star (BNS) mergers, with the help of early warnings from decihertz gravitational-wave (GW) observatories, B-DECIGO and DO-Optimal. Extending the previous work, we not only give quick assessments of joint short $γ$-ray burst (sGRB) detection rates for different $γ$-ray satellites and BNS population models, but also elaborate on the analyses and results on multi-band kilonova detections for survey telescopes with different limiting magnitudes. During an assumed 4-year mission time for decihertz GW observatories, we find that for the goals of electromagnetic follow-ups, DO-Optimal performs better than B-DECIGO as a whole on the detection rate, and has a larger detectable distance for joint sGRB/kilonova searches. Taking the log-normal population model for BNS mergers and a one-day early-warning time as an example, we discuss the accuracy in localization and timing, as well as the redshift distributions for various synergy observations with electromagnetic facilities and decihertz GW detectors. Based on our analyses, we propose a feasible "wait-for" pattern as a novel detecting mode for future multi-messenger astrophysics.

preprint2022arXiv

Extending the Fisher Information Matrix in Gravitational-wave Data Analysis

The Fisher information matrix (FM) plays an important role in forecasts and inferences in many areas of physics. While giving fast parameter estimation with the Gaussian likelihood approximation in the parameter space, the FM can only give the ellipsoidal posterior contours of parameters and lose the higher-order information beyond Gaussianity. We extend the FM in gravitational-wave (GW) data analysis using the Derivative Approximation for LIkelihoods (DALI), a method to expand the likelihood while keeping it positive definite and normalizable at every order, for more accurate forecasts and inferences. When applied to the two real GW events, GW150914 and GW170817, DALI can reduce the difference between FM approximation and the real posterior by 5 times in the best case. The calculation time of DALI and FM is at the same order of magnitude, while obtaining the real full posterior will take several orders of magnitude longer. Besides more accurate approximations, higher-order correction from DALI provides a fast assessment on the FM analysis and gives suggestions for complex sampling techniques which are computationally intensive. We recommend using the DALI method as an extension to the FM method in GW data analysis to pursue better accuracy while still keeping the speed.

preprint2022arXiv

Generalizing to Unseen Domains: A Survey on Domain Generalization

Machine learning systems generally assume that the training and testing distributions are the same. To this end, a key requirement is to develop models that can generalize to unseen distributions. Domain generalization (DG), i.e., out-of-distribution generalization, has attracted increasing interests in recent years. Domain generalization deals with a challenging setting where one or several different but related domain(s) are given, and the goal is to learn a model that can generalize to an unseen test domain. Great progress has been made in the area of domain generalization for years. This paper presents the first review of recent advances in this area. First, we provide a formal definition of domain generalization and discuss several related fields. We then thoroughly review the theories related to domain generalization and carefully analyze the theory behind generalization. We categorize recent algorithms into three classes: data manipulation, representation learning, and learning strategy, and present several popular algorithms in detail for each category. Third, we introduce the commonly used datasets, applications, and our open-sourced codebase for fair evaluation. Finally, we summarize existing literature and present some potential research topics for the future.

preprint2022arXiv

Hot-tail electrons' impact on assimilation and injection penetration of D2 Shattered Pellet Injections

The fragment ablation rate plays significant roles in the mitigation efficiency of Shattered Pellet Injection (SPI) as a Disruption Mitigation System (DMS). Current mainstream 3D MHD codes modelling SPIs mostly assume instantaneous thermalization between the previously hot ambient electrons and the newly released cold electrons, which results in underestimation of the ablation rate if the hot electron thermalization time is comparable or even longer than the fragment flying time. To resolve this doubt, we hereby investigate the thermalization dynamics and the overall hot-electron impact. The finite-time collisional thermalization of hot-tail electrons in a rapidly cooling plasma, as well as the so-called ``self-limiting'' effect are considered. The former effect tends to deplete the colder population within a hot-tail species, while the latter is found to preferentially deplete the higher energy population. The combined result is found to cause an almost self-similar decay of the hot electron distribution function, while its shape does not deviate much from that of Maxwellian distribution and the mean energy does not change much during the thermalization process. Based on this observation, axisymmetric JOREK D2 SPI simulations were carried out with additional hot-tail contribution to evaluate their overall impact onto the injection assimilation and penetration. It is found that the hot-tail effect indeed causes enhanced assimilation and shallower penetration, although the overall effect depends on the exact injection configuration, with the slow injection showing negligible hot-tail effect while the fast single non-shattered pellet case shows drastic hot-tail ablation enhancement. For ITER-like SPI parameters, there is no significant deviation in the total assimilation, but some deviation in the injection penetration is observed for the fast injection velocity cases.

preprint2022arXiv

Improving the Robustness of Adversarial Attacks Using an Affine-Invariant Gradient Estimator

As designers of artificial intelligence try to outwit hackers, both sides continue to hone in on AI's inherent vulnerabilities. Designed and trained from certain statistical distributions of data, AI's deep neural networks (DNNs) remain vulnerable to deceptive inputs that violate a DNN's statistical, predictive assumptions. Before being fed into a neural network, however, most existing adversarial examples cannot maintain malicious functionality when applied to an affine transformation. For practical purposes, maintaining that malicious functionality serves as an important measure of the robustness of adversarial attacks. To help DNNs learn to defend themselves more thoroughly against attacks, we propose an affine-invariant adversarial attack, which can consistently produce more robust adversarial examples over affine transformations. For efficiency, we propose to disentangle current affine-transformation strategies from the Euclidean geometry coordinate plane with its geometric translations, rotations and dilations; we reformulate the latter two in polar coordinates. Afterwards, we construct an affine-invariant gradient estimator by convolving the gradient at the original image with derived kernels, which can be integrated with any gradient-based attack methods. Extensive experiments on ImageNet, including some experiments under physical condition, demonstrate that our method can significantly improve the affine invariance of adversarial examples and, as a byproduct, improve the transferability of adversarial examples, compared with alternative state-of-the-art methods.

preprint2022arXiv

Increased Phase Coherence Length in a Porous Topological Insulator

The surface area of Bi2Te3 thin films was increased by introducing nanoscale porosity. Temperature dependent resistivity and magnetotransport measurements were conducted both on as-grown and porous samples (23 and 70 nm). The longitudinal resistivity of the porous samples became more metallic, indicating the increased surface area resulted in transport that was more surface-like. Weak antilocalization (WAL) was present in all samples, and remarkably the phase coherence length doubled in the porous samples. This increase is likely due to the large Fermi velocity of the Dirac surface states. Our results show that the introduction of nanoporosity does not destroy the topological surface states but rather enhances them, making these nanostructured materials promising for low energy electronics, spintronics and thermoelectrics.

preprint2022arXiv

Influences of the dissipative topological edge state on quantized transport in MnBi2Te4

The beauty of quantum Hall (QH) effect is the metrological precision of Hall resistance quantization that originates from the topological edge states. Understanding the factors that lead to quantization breakdown not only provides important insights on the nature of the topological protection of these edge states, but is beneficial for device applications involving such quantized transport. In this work, we combine conventional transport and real space conductivity mapping to investigate whether the quantization breakdown is tied to the disappearance of edge state in the hotly studied MnBi2Te4 system. Our experimental results unambiguously show that topological edge state does exist when quantization breakdown occurs. Such edge state is dissipative in nature and could lead to a quantization breakdown due to its diffusive character causing overlapping with bulk and other edge states in real devices. Our findings bring attentions to issues that are generally inaccessible in the transport study of QH, but can play important roles in practical measurements and device applications.

preprint2022arXiv

Infrared Radiation of Graphene Electrothermal Film Triggered Alpha and Theta Brainwaves

The alpha and theta frequency brainwave activity in Electroencephalogram (EEG) signal has been correlated with attention, inhibitory processes, memory, perceptual abilities, and sleep. The enhanced alpha and theta brainwave activity may bring positive behavioral modifications such as promoting creativity and a quick sleep. Herein, we discover that infrared radiation from multilayer graphene electrothermal film can obviously promote the appearance of alpha and theta brainwave in human mind. In particular, the occurrence frequency of the alpha and theta waves in EEG can be effectively enhanced up to 2.3 and 3.0 times, respectively. And the duration time of the alpha and theta waves in EEG can also be effectively extended. The mechanism may be attributed to the efficient infrared radiation caused by graphene mainly focused on the range from 7 to 14 micron, coinciding with the radiation wavelength of natural human body, which can be effectively absorbed by the human skin and speed up the blood microcirculation and metabolism. The comparative effect of different working temperature and heating materials such as water, Cu and even monolayer graphene are systematically investigated, indicating the infrared radiation from the multilayer graphene electrothermal film at 50 degrees has the largest enhancement effect of alpha and theta brainwaves. The multilayer graphene film electrical heater represents a convenient and surprising way for triggering the alpha and theta brainwaves, which has many potential applications in the area of enlarged health cerements.

preprint2022arXiv

Instance-Specific Feature Propagation for Referring Segmentation

Referring segmentation aims to generate a segmentation mask for the target instance indicated by a natural language expression. There are typically two kinds of existing methods: one-stage methods that directly perform segmentation on the fused vision and language features; and two-stage methods that first utilize an instance segmentation model for instance proposal and then select one of these instances via matching them with language features. In this work, we propose a novel framework that simultaneously detects the target-of-interest via feature propagation and generates a fine-grained segmentation mask. In our framework, each instance is represented by an Instance-Specific Feature (ISF), and the target-of-referring is identified by exchanging information among all ISFs using our proposed Feature Propagation Module (FPM). Our instance-aware approach learns the relationship among all objects, which helps to better locate the target-of-interest than one-stage methods. Comparing to two-stage methods, our approach collaboratively and interactively utilizes both vision and language information for synchronous identification and segmentation. In the experimental tests, our method outperforms previous state-of-the-art methods on all three RefCOCO series datasets.

preprint2022arXiv

Learning-based Predictive Beamforming for Integrated Sensing and Communication in Vehicular Networks

This paper investigates the integrated sensing and communication (ISAC) in vehicle-to-infrastructure (V2I) networks. To realize ISAC, an effective beamforming design is essential which however, highly depends on the availability of accurate channel tracking requiring large training overhead and computational complexity. Motivated by this, we adopt a deep learning (DL) approach to implicitly learn the features of historical channels and directly predict the beamforming matrix to be adopted for the next time slot to maximize the average achievable sum-rate of an ISAC system. The proposed method can bypass the need of explicit channel tracking process and reduce the signaling overhead significantly. To this end, a general sum-rate maximization problem with Cramer-Rao lower bounds (CRLBs)-based sensing constraints is first formulated for the considered ISAC system taking into account the multiple access interference. Then, by exploiting the penalty method, a versatile unsupervised DL-based predictive beamforming design framework is developed to address the formulated design problem. As a realization of the developed framework, a historical channels-based convolutional long short-term memory (LSTM) network (HCL-Net) is devised for predictive beamforming in the ISAC-based V2I network. Specifically, the convolution and LSTM modules are successively adopted in the proposed HCL-Net to exploit the spatial and temporal dependencies of communication channels to further improve the learning performance. Finally, simulation results show that the proposed predictive method not only guarantees the required sensing performance, but also achieves a satisfactory sum-rate that can approach the upper bound obtained by the genie-aided scheme with the perfect instantaneous channel state information available.

preprint2022arXiv

MAP-Gen: An Automated 3D-Box Annotation Flow with Multimodal Attention Point Generator

Manually annotating 3D point clouds is laborious and costly, limiting the training data preparation for deep learning in real-world object detection. While a few previous studies tried to automatically generate 3D bounding boxes from weak labels such as 2D boxes, the quality is sub-optimal compared to human annotators. This work proposes a novel autolabeler, called multimodal attention point generator (MAP-Gen), that generates high-quality 3D labels from weak 2D boxes. It leverages dense image information to tackle the sparsity issue of 3D point clouds, thus improving label quality. For each 2D pixel, MAP-Gen predicts its corresponding 3D coordinates by referencing context points based on their 2D semantic or geometric relationships. The generated 3D points densify the original sparse point clouds, followed by an encoder to regress 3D bounding boxes. Using MAP-Gen, object detection networks that are weakly supervised by 2D boxes can achieve 94~99% performance of those fully supervised by 3D annotations. It is hopeful this newly proposed MAP-Gen autolabeling flow can shed new light on utilizing multimodal information for enriching sparse point clouds.

preprint2022arXiv

Medical Knowledge-Guided Deep Learning for Imbalanced Medical Image Classification

Deep learning models have gained remarkable performance on a variety of image classification tasks. However, many models suffer from limited performance in clinical or medical settings when data are imbalanced. To address this challenge, we propose a medical-knowledge-guided one-class classification approach that leverages domain-specific knowledge of classification tasks to boost the model's performance. The rationale behind our approach is that some existing prior medical knowledge can be incorporated into data-driven deep learning to facilitate model learning. We design a deep learning-based one-class classification pipeline for imbalanced image classification, and demonstrate in three use cases how we take advantage of medical knowledge of each specific classification task by generating additional middle classes to achieve higher classification performances. We evaluate our approach on three different clinical image classification tasks (a total of 8459 images) and show superior model performance when compared to six state-of-the-art methods. All codes of this work will be publicly available upon acceptance of the paper.

preprint2022arXiv

Multimodal Transformer for Automatic 3D Annotation and Object Detection

Despite a growing number of datasets being collected for training 3D object detection models, significant human effort is still required to annotate 3D boxes on LiDAR scans. To automate the annotation and facilitate the production of various customized datasets, we propose an end-to-end multimodal transformer (MTrans) autolabeler, which leverages both LiDAR scans and images to generate precise 3D box annotations from weak 2D bounding boxes. To alleviate the pervasive sparsity problem that hinders existing autolabelers, MTrans densifies the sparse point clouds by generating new 3D points based on 2D image information. With a multi-task design, MTrans segments the foreground/background, densifies LiDAR point clouds, and regresses 3D boxes simultaneously. Experimental results verify the effectiveness of the MTrans for improving the quality of the generated labels. By enriching the sparse point clouds, our method achieves 4.48\% and 4.03\% better 3D AP on KITTI moderate and hard samples, respectively, versus the state-of-the-art autolabeler. MTrans can also be extended to improve the accuracy for 3D object detection, resulting in a remarkable 89.45\% AP on KITTI hard samples. Codes are at \url{https://github.com/Cliu2/MTrans}.

preprint2022arXiv

Nonlinear inhomogeneous Fokker-Planck models: energetic-variational structures and long time behavior

Inspired by the modeling of grain growth in polycrystalline materials, we consider a nonlinear Fokker-Plank model, with inhomogeneous diffusion and with variable mobility parameters. We develop large time asymptotic analysis of such nonstandard models by reformulating and extending the classical entropy method, under the assumption of periodic boundary condition. In addition, illustrative numerical tests are presented to highlight the essential points of the current analytical results and to motivate future analysis.

preprint2022arXiv

NTIRE 2022 Challenge on High Dynamic Range Imaging: Methods and Results

This paper reviews the challenge on constrained high dynamic range (HDR) imaging that was part of the New Trends in Image Restoration and Enhancement (NTIRE) workshop, held in conjunction with CVPR 2022. This manuscript focuses on the competition set-up, datasets, the proposed methods and their results. The challenge aims at estimating an HDR image from multiple respective low dynamic range (LDR) observations, which might suffer from under- or over-exposed regions and different sources of noise. The challenge is composed of two tracks with an emphasis on fidelity and complexity constraints: In Track 1, participants are asked to optimize objective fidelity scores while imposing a low-complexity constraint (i.e. solutions can not exceed a given number of operations). In Track 2, participants are asked to minimize the complexity of their solutions while imposing a constraint on fidelity scores (i.e. solutions are required to obtain a higher fidelity score than the prescribed baseline). Both tracks use the same data and metrics: Fidelity is measured by means of PSNR with respect to a ground-truth HDR image (computed both directly and with a canonical tonemapping operation), while complexity metrics include the number of Multiply-Accumulate (MAC) operations and runtime (in seconds).

preprint2022arXiv

On the $α$-spectral radius of the $k$-uniform supertrees

Let $G$ be a $k$-uniform hypergraph with vertex set $V(G)$ and edge set $E(G)$. A connected and acyclic hypergraph is called a supertree. For $0\leqα<1$, the $α$-spectral radius of $G$ is the largest $H$-eigenvalue of $αD(G)+(1-α)A(G)$, where $D(G)$ and $A(G)$ are the diagonal tensor of the degrees and the adjacency tensor of $G$, respectively. In this paper, we determine the unique supertrees with the maximum $α$-spectral radius among all $k$-uniform supertrees with $m$ edges and independence number $β$ for $\lceil\frac{m(k-1)+1}{k}\rceil\leqβ\leq m$, among all $k$-uniform supertrees with given degree sequences, and among all $k$-uniform supertrees with $m$ edges and matching number $μ$ for $1\leqμ\leq\lfloor\frac{m(k-1)+1}{k}\rfloor$, respectively.

preprint2022arXiv

Open-set Text Recognition via Character-Context Decoupling

The open-set text recognition task is an emerging challenge that requires an extra capability to cognize novel characters during evaluation. We argue that a major cause of the limited performance for current methods is the confounding effect of contextual information over the visual information of individual characters. Under open-set scenarios, the intractable bias in contextual information can be passed down to visual information, consequently impairing the classification performance. In this paper, a Character-Context Decoupling framework is proposed to alleviate this problem by separating contextual information and character-visual information. Contextual information can be decomposed into temporal information and linguistic information. Here, temporal information that models character order and word length is isolated with a detached temporal attention module. Linguistic information that models n-gram and other linguistic statistics is separated with a decoupled context anchor mechanism. A variety of quantitative and qualitative experiments show that our method achieves promising performance on open-set, zero-shot, and close-set text recognition datasets.

preprint2022arXiv

Orientation-Discriminative Feature Representation for Decentralized Pedestrian Tracking

This paper focuses on the problem of decentralized pedestrian tracking using a sensor network. Traditional works on pedestrian tracking usually use a centralized framework, which becomes less practical for robotic applications due to limited communication bandwidth. Our paper proposes a communication-efficient, orientation-discriminative feature representation to characterize pedestrian appearance information, that can be shared among sensors. Building upon that representation, our work develops a cross-sensor track association approach to achieve decentralized tracking. Extensive evaluations are conducted on publicly available datasets and results show that our proposed approach leads to improved performance in multi-sensor tracking.

preprint2022arXiv

Predictive Beamforming for Integrated Sensing and Communication in Vehicular Networks: A Deep Learning Approach

The implementation of integrated sensing and communication (ISAC) highly depends on the effective beamforming design exploiting accurate instantaneous channel state information (ICSI). However, channel tracking in ISAC requires large amount of training overhead and prohibitively large computational complexity. To address this problem, in this paper, we focus on ISAC-assisted vehicular networks and exploit a deep learning approach to implicitly learn the features of historical channels and directly predict the beamforming matrix for the next time slot to maximize the average achievable sum-rate of system, thus bypassing the need of explicit channel tracking for reducing the system signaling overhead. To this end, a general sum-rate maximization problem with Cramer-Rao lower bounds-based sensing constraints is first formulated for the considered ISAC system. Then, a historical channels-based convolutional long short-term memory network is designed for predictive beamforming that can exploit the spatial and temporal dependencies of communication channels to further improve the learning performance. Finally, simulation results show that the proposed method can satisfy the requirement of sensing performance, while its achievable sum-rate can approach the upper bound obtained by a genie-aided scheme with perfect ICSI available.

preprint2022arXiv

Preferred Spin Excitations in the Bilayer Iron-Based Superconductor CaK(Fe$_{0.96}$Ni$_{0.04}$)$_4$As$_4$ with Spin-Vortex Crystal Order

The spin-orbit coupling (SOC) is a key to understand the magnetically driven superconductivity in iron-based superconductors, where both local and itinerant electrons are present and the orbital angular momentum is not completely quenched. Here, we report a neutron scattering study on the bilayer compound CaK(Fe$_{0.96}$Ni$_{0.04}$)$_4$As$_4$ with superconductivity coexisting with a non-collinear spin-vortex crystal magnetic order that preserves the tetragonal symmetry of Fe-Fe plane. In the superconducting state, two spin resonance modes with odd and even $L$ symmetries due to the bilayer coupling are found similar to the undoped compound CaKFe$_4$As$_4$ but at lower energies. Polarization analysis reveals that the odd mode is $c-$axis polarized, and the low-energy spin anisotropy can persist to the paramagnetic phase at high temperature, which closely resembles other systems with in-plane collinear and $c-$axis biaxial magnetic orders. These results provide the missing piece of the puzzle on the SOC effect in iron-pnictide superconductors, and also establish a common picture of $c-$axis preferred magnetic excitations below $T_c$ regardless of the details of magnetic pattern or lattice symmetry.

preprint2022arXiv

Primal-dual Estimator Learning: an Offline Constrained Moving Horizon Estimation Method with Feasibility and Near-optimality Guarantees

This paper proposes a primal-dual framework to learn a stable estimator for linear constrained estimation problems leveraging the moving horizon approach. To avoid the online computational burden in most existing methods, we learn a parameterized function offline to approximate the primal estimate. Meanwhile, a dual estimator is trained to check the suboptimality of the primal estimator during execution time. Both the primal and dual estimators are learned from data using supervised learning techniques, and the explicit sample size is provided, which enables us to guarantee the quality of each learned estimator in terms of feasibility and optimality. This in turn allows us to bound the probability of the learned estimator being infeasible or suboptimal. Furthermore, we analyze the stability of the resulting estimator with a bounded error in the minimization of the cost function. Since our algorithm does not require the solution of an optimization problem during runtime, state estimates can be generated online almost instantly. Simulation results are presented to show the accuracy and time efficiency of the proposed framework compared to online optimization of moving horizon estimation and Kalman filter. To the best of our knowledge, this is the first learning-based state estimator with feasibility and near-optimality guarantees for linear constrained systems.

preprint2022arXiv

Pronunciation Dictionary-Free Multilingual Speech Synthesis by Combining Unsupervised and Supervised Phonetic Representations

This paper proposes a multilingual speech synthesis method which combines unsupervised phonetic representations (UPR) and supervised phonetic representations (SPR) to avoid reliance on the pronunciation dictionaries of target languages. In this method, a pretrained wav2vec 2.0 model is adopted to extract UPRs and a language-independent automatic speech recognition (LI-ASR) model is built with a connectionist temporal classification (CTC) loss to extract segment-level SPRs from the audio data of target languages. Then, an acoustic model is designed, which first predicts UPRs and SPRs from texts separately and then combines the predicted UPRs and SPRs to generate mel-spectrograms. The results of our experiments on six languages show that the proposed method outperformed the methods that directly predicted mel-spectrograms from character or phoneme sequences and the ablated models that utilized only UPRs or SPRs.

preprint2022arXiv

Realistic Detection and Early Warning of Binary Neutron Stars with Decihertz Gravitational-wave Observatories

We investigated the detection rates and early warning parameters of binary neutron star (BNS) populations with decihertz gravitational-wave observatories in a realistic detecting strategy. Assuming 4 years' operation of B-DECIGO, we based on parameter precision to classify the detectable BNSs into three categories: (a) sources that merge within 1 year, which could be localized with an uncertainty of $ΔΩ\sim 10^{0}$ deg$^2$; (b) sources that merge in 1-4 years, which take up three quarters of the total events and yield the most precise angular resolution with $ΔΩ\sim 10^{-2}$ deg$^2$ and time-of-merger accuracy with $Δt_c\sim 10^{-1}$ s; and (c) sources that do not merge during the 4-yr mission window, which enable possible early warnings, with $ΔΩ\sim 10^{-1}$ deg$^2$ and $Δt_c\sim 10^{0}$ s. Furthermore, we compared the pros and cons of B-DECIGO with the third-generation ground-based detectors, and explored the prospects of detections using 3 other decihertz observatories and 4 BNS population models. In realistic observing scenarios, we found that decihertz detectors could even provide early-warning alerts to a source decades before its merger while their localizations are still as accurate as ground-based facilities. Finally we found a decrease of events when considering the confusion noise, but this could be partially solved by a proper noise subtraction.

preprint2022arXiv

Robust quantum control for the manipulation of solid-state spins

Robust and high-fidelity control of electron spins in solids is the cornerstone for facilitating applications of solid-state spins in quantum information processing and quantum sensing. However, precise control of spin systems is always challenging due to the presence of a variety of noises originating from the environment and control fields. Herein, noise-resilient quantum gates, designed with robust optimal control (ROC) algorithms, are demonstrated experimentally with nitrogen-vacancy (NV) centers in diamond to realize tailored robustness against detunings and Rabi errors simultaneously. In the presence of both 10% off-resonant detuning and deviation of a Rabi frequency, we achieve an average single-qubit gate fidelity of up to 99.97%. Our experiments also show that, ROCbased multipulse quantum sensing sequences can suppress spurious responses resulting from finite widths and imperfections of microwave pulses, which provides an efficient strategy for enhancing the performance of existing multipulse quantum sensing sequences.

preprint2022arXiv

Sampling with Mirrored Stein Operators

We introduce a new family of particle evolution samplers suitable for constrained domains and non-Euclidean geometries. Stein Variational Mirror Descent and Mirrored Stein Variational Gradient Descent minimize the Kullback-Leibler (KL) divergence to constrained target distributions by evolving particles in a dual space defined by a mirror map. Stein Variational Natural Gradient exploits non-Euclidean geometry to more efficiently minimize the KL divergence to unconstrained targets. We derive these samplers from a new class of mirrored Stein operators and adaptive kernels developed in this work. We demonstrate that these new samplers yield accurate approximations to distributions on the simplex, deliver valid confidence intervals in post-selection inference, and converge more rapidly than prior methods in large-scale unconstrained posterior inference. Finally, we establish the convergence of our new procedures under verifiable conditions on the target distribution.

preprint2022arXiv

Self-driven water or polarized liquid based ultraviolet photodetector

Traditionally, photodetector is based on solid materials constructed PN junction, which needs many delicate growth technologies. Herein, we demonstrate the feasibility of polarized liquid triggered photodetector where the liquid is sandwiched between P-type or N-type semiconductor which can be chosen freely according to the requirement of the specific response wavelength. Under the multiple cycles of optical switching, transient photo-polarized current, steady state photo-polarized current and depolarized current are repeatably observed in semiconductor/polar liquids/semiconductor structure. The responsivity and specific detectivity of transient photo-polarized current in N-GaN/water/P-GaN reach values of 104.2 mA/W and 4.4*10^12 Jones at 365 nm, and 52.8 mA/W and 1.9*10^12 Jones at 254 nm illumination under zero voltage bias. We anticipate that our research will have a profound impact on integrating self-powered photodetector with freely selectable wavelength bands.

preprint2022arXiv

SiamHAN: IPv6 Address Correlation Attacks on TLS Encrypted Traffic via Siamese Heterogeneous Graph Attention Network

Unlike IPv4 addresses, which are typically masked by a NAT, IPv6 addresses could easily be correlated with user activity, endangering their privacy. Mitigations to address this privacy concern have been deployed, making existing approaches for address-to-user correlation unreliable. This work demonstrates that an adversary could still correlate IPv6 addresses with users accurately, even with these protection mechanisms. To do this, we propose an IPv6 address correlation model - SiamHAN. The model uses a Siamese Heterogeneous Graph Attention Network to measure whether two IPv6 client addresses belong to the same user even if the user's traffic is protected by TLS encryption. Using a large real-world dataset, we show that, for the tasks of tracking target users and discovering unique users, the state-of-the-art techniques could achieve only 85% and 60% accuracy, respectively. However, SiamHAN exhibits 99% and 88% accuracy.

preprint2022arXiv

Spatial input-output analysis of large-scale structures in actuated turbulent boundary layers

This paper develops a spatial input-output approach to investigate the dynamics of a turbulent boundary layer subject to a localized single frequency excitation. This method uses one-way spatial integration to reformulate the problem in terms of spatial evolution equations. The technique is used to examine the effect of localized periodic actuation at a given temporal frequency, based on an experimental set-up in which an active large-scale is introduced into the outer layer of a turbulent boundary layer. First, the large-scale structures associated with the phase-locked modal velocity field obtained from spatial input-output analysis are shown to closely match those computed based on hot-wire measurements. The approach is then used to further investigate the response of the boundary layer to the synthetically generated large-scale. A quadrant trajectory analysis indicates that the spatial input-output response produces shear stress distributions consistent with those in canonical wall-bounded turbulent flows in terms of both the order and types of events observed. The expected correspondence between the dominance of different quadrant behavior and actuation frequency is also observed. These results highlight the promise of a spatial input-output framework for analyzing the formation and streamwise evolution of structures in actuated wall-bounded turbulent flows.

preprint2022arXiv

Strain-driven valley states and phase transitions in Janus VSiGeN4 monolayer

The interplay between topology and valley degree of freedom has attracted much interest because it can realize new phenomena and applications. Here, based on first-principles calculations, we demonstrate intrinsically valley-polarized quantum anomalous Hall effect in monolayer ferrovalley material: Janus VSiGeN4, of which the edge states are chiral-spin-valley locking. Besides, a small tensile or compressive strain can drive phase transition in the material from valley-polarized quantum anomalous Hall state to half-valley-metal state. With the increase of the strain, the material turns into ferrovalley semiconductor with valley anomalous Hall effect. The origin of phase transition is sequent band inversion of V d orbital at K valley. Moreover, we find that phase transition causes the sign reversal of Berry curvature and induces different polarized light absorption in different valley states. Our work provides an ideal material platform for practical applications and experimental exploration of the interplay between topology, spintronics, and valleytronics.

preprint2022arXiv

TANet: Transformer-based Asymmetric Network for RGB-D Salient Object Detection

Existing RGB-D SOD methods mainly rely on a symmetric two-stream CNN-based network to extract RGB and depth channel features separately. However, there are two problems with the symmetric conventional network structure: first, the ability of CNN in learning global contexts is limited; second, the symmetric two-stream structure ignores the inherent differences between modalities. In this paper, we propose a Transformer-based asymmetric network (TANet) to tackle the issues mentioned above. We employ the powerful feature extraction capability of Transformer (PVTv2) to extract global semantic information from RGB data and design a lightweight CNN backbone (LWDepthNet) to extract spatial structure information from depth data without pre-training. The asymmetric hybrid encoder (AHE) effectively reduces the number of parameters in the model while increasing speed without sacrificing performance. Then, we design a cross-modal feature fusion module (CMFFM), which enhances and fuses RGB and depth features with each other. Finally, we add edge prediction as an auxiliary task and propose an edge enhancement module (EEM) to generate sharper contours. Extensive experiments demonstrate that our method achieves superior performance over 14 state-of-the-art RGB-D methods on six public datasets. Our code will be released at https://github.com/lc012463/TANet.

preprint2022arXiv

Test-time Fourier Style Calibration for Domain Generalization

The topic of generalizing machine learning models learned on a collection of source domains to unknown target domains is challenging. While many domain generalization (DG) methods have achieved promising results, they primarily rely on the source domains at train-time without manipulating the target domains at test-time. Thus, it is still possible that those methods can overfit to source domains and perform poorly on target domains. Driven by the observation that domains are strongly related to styles, we argue that reducing the gap between source and target styles can boost models' generalizability. To solve the dilemma of having no access to the target domain during training, we introduce Test-time Fourier Style Calibration (TF-Cal) for calibrating the target domain style on the fly during testing. To access styles, we utilize Fourier transformation to decompose features into amplitude (style) features and phase (semantic) features. Furthermore, we present an effective technique to Augment Amplitude Features (AAF) to complement TF-Cal. Extensive experiments on several popular DG benchmarks and a segmentation dataset for medical images demonstrate that our method outperforms state-of-the-art methods.

preprint2022arXiv

The maximum $A_α$-spectral radius of $t$-connected graphs with bounded matching number

Let $G$ be a graph with adjacency matrix $A(G)$ and let $D(G)$ be a diagonal matrix of the degrees of $G$. In 2017, Nikiforov defined the $A_α$-matrix of $G$ as \begin{equation*} A_α(G)=αG)+(1-α)A(G), \end{equation*}d where $α\in[0,1]$ is an arbitrary real number. The largest eigenvalue of $A_α(G)$ is called the $A_α$-spectral radius of $G$. Let $n$, $t$, $k$ be positive integers, satisfying $t\geq1$, $k\geq2$, $n\geq k+2$, and $n\equiv k$ (mod $2$). In this paper, for $α\in[0,\frac{1}{2}]$, we determine the extremal graphs with the maximum $A_α$-spectral radius among all $t$-connected graphs on $n$ vertices with matching number $\frac{n-k}{2}$ at most. This generalizes some results of O (2021) and Zhang (2022).

preprint2022arXiv

Time-Frequency Mask Aware Bi-directional LSTM: A Deep Learning Approach for Underwater Acoustic Signal Separation

The underwater acoustic signals separation is a key technique for the underwater communications. The existing methods are mostly model-based, and could not accurately characterise the practical underwater acoustic communication environment. They are only suitable for binary signal separation, but cannot handle multivariate signal separation. On the other hand, the recurrent neural network (RNN) shows powerful capability in extracting the features of the temporal sequences. Inspired by this, in this paper, we present a data-driven approach for underwater acoustic signals separation using deep learning technology. We use the Bi-directional Long Short-Term Memory (Bi-LSTM) to explore the features of Time-Frequency (T-F) mask, and propose a T-F mask aware Bi-LSTM for signal separation. Taking advantage of the sparseness of the T-F image, the designed Bi-LSTM network is able to extract the discriminative features for separation, which further improves the separation performance. In particular, this method breaks through the limitations of the existing methods, not only achieves good results in multivariate separation, but also effectively separates signals when mixed with 40dB Gaussian noise signals. The experimental results show that this method can achieve a $97\%$ guarantee ratio (PSR), and the average similarity coefficient of the multivariate signal separation is stable above 0.8 under high noise conditions.

preprint2022arXiv

Topology optimization on complex surfaces based on the moving morphable component (MMC) method and computational conformal mapping (CCM)

In the present paper, an integrated paradigm for topology optimization on complex surfaces with arbitrary genus is proposed. The approach is constructed based on the two-dimensional (2D) Moving Morphable Component (MMC) framework, where a set of structural components are used as the basic units of optimization, and computational conformal mapping (CCM) technique, with which a complex surface represented by an unstructured triangular mesh can be mapped into a set of regular 2D parameter domains numerically. A multi-patch stitching scheme is also developed to achieve an MMC-friendly global parameterization through a number of local parameterizations. Numerical examples including a saddle-shaped shell, a torus-shape shell and a tee-branch pipe are solved to demonstrate the validity and efficiency of the proposed approach. It is found that compared with traditional approaches for topology optimization on 2D surfaces, optimized designs with clear load transmission paths can be obtained with much fewer numbers of design variables and degrees of freedom for finite element analysis (FEA) via the proposed approach.

preprint2022arXiv

Towards Open-Set Text Recognition via Label-to-Prototype Learning

Scene text recognition is a popular topic and extensively used in the industry. Although many methods have achieved satisfactory performance for the close-set text recognition challenges, these methods lose feasibility in open-set scenarios, where collecting data or retraining models for novel characters could yield a high cost. For example, annotating samples for foreign languages can be expensive, whereas retraining the model each time when a novel character is discovered from historical documents costs both time and resources. In this paper, we introduce and formulate a new open-set text recognition task which demands the capability to spot and recognize novel characters without retraining. A label-to-prototype learning framework is also proposed as a baseline for the proposed task. Specifically, the framework introduces a generalizable label-to-prototype mapping function to build prototypes (class centers) for both seen and unseen classes. An open-set predictor is then utilized to recognize or reject samples according to the prototypes. The implementation of rejection capability over out-of-set characters allows automatic spotting of unknown characters in the incoming data stream. Extensive experiments show that our method achieves promising performance on a variety of zero-shot, close-set, and open-set text recognition datasets

preprint2022arXiv

Tri-branched gels: Rubbery materials with the lowest branching factor approach the ideal elastic limit

Unlike hard materials such as metals and ceramics, rubbery materials can endure large deformations due to the large conformational degree of freedom of the crosslinked three-dimensional polymer network. However, the effect of the branching factor of the network on the ultimate mechanical properties of rubbery materials has not yet been clarified. This study shows that tri-branching, which entails the lowest branching factor, results in a large elastic deformation near the theoretical upper bound. This ideal elastic limit is realized by reversible strain-induced crystallization, providing on-demand reinforcement. The findings indicate that the polymer chain is highly orientated along the stretching axis, whereat enhanced reversible strain-induced crystallization is observed in the tri-branched and not in the tetra-branched network. A mathematical theory of structural rigidity is used to explain the difference in the chain orientation. Although tetra-branched polymers have been preferred since the development of vulcanization, these findings highlighting the merits of tri-branching will prompt a paradigm shift in the development of rubbery materials.

preprint2022arXiv

TTAGN: Temporal Transaction Aggregation Graph Network for Ethereum Phishing Scams Detection

In recent years, phishing scams have become the most serious type of crime involved in Ethereum, the second-largest blockchain platform. The existing phishing scams detection technology on Ethereum mostly uses traditional machine learning or network representation learning to mine the key information from the transaction network to identify phishing addresses. However, these methods adopt the last transaction record or even completely ignore these records, and only manual-designed features are taken for the node representation. In this paper, we propose a Temporal Transaction Aggregation Graph Network (TTAGN) to enhance phishing scams detection performance on Ethereum. Specifically, in the temporal edges representation module, we model the temporal relationship of historical transaction records between nodes to construct the edge representation of the Ethereum transaction network. Moreover, the edge representations around the node are aggregated to fuse topological interactive relationships into its representation, also named as trading features, in the edge2node module. We further combine trading features with common statistical and structural features obtained by graph neural networks to identify phishing addresses. Evaluated on real-world Ethereum phishing scams datasets, our TTAGN (92.8% AUC, and 81.6% F1score) outperforms the state-of-the-art methods, and the effectiveness of temporal edges representation and edge2node module is also demonstrated.

preprint2022arXiv

Variational Policy Propagation for Multi-agent Reinforcement Learning

We propose a \emph{collaborative} multi-agent reinforcement learning algorithm named variational policy propagation (VPP) to learn a \emph{joint} policy through the interactions over agents. We prove that the joint policy is a Markov Random Field under some mild conditions, which in turn reduces the policy space effectively. We integrate the variational inference as special differentiable layers in policy such that the actions can be efficiently sampled from the Markov Random Field and the overall policy is differentiable. We evaluate our algorithm on several large scale challenging tasks and demonstrate that it outperforms previous state-of-the-arts.

preprint2021arXiv

A Computational Design and Evaluation Tool for 3D Structures with Planar Surfaces

Three dimensional (3D) structures composed of planar surfaces can be build out of accessible materials using easier fabrication technique with shorter fabrication time. To better design 3D structures with planar surfaces, realistic models are required to understand and evaluate mechanical behaviors. Existing design tools are either effort-consuming (e.g. finite element analysis) or bounded by assumptions (e.g. numerical solutions). In this project, We have built a computational design tool that is (1) capable of rapidly and inexpensively evaluating planar surfaces in 3D structures, with sufficient computational efficiency and accuracy; (2) applicable to complex boundary conditions and loading conditions, both isotropic materials and orthotropic materials; and (3) suitable for rapid accommodation when design parameters need to be adjusted. We demonstrate the efficiency and necessity of this design tool by evaluating a glass table as well as a wood bookcase, and iteratively designing an origami gripper to satisfy performance requirements. This design tool gives non-expert users as well as engineers a simple and effective modus operandi in structural design.

preprint2021arXiv

Ambi-chiral anomalous Hall effect in magnetically doped topological insulators

The chirality associated with broken time reversal symmetry in magnetically doped topological insulators has important implications to the quantum transport phenomena. Here we report the anomalous Hall effect studies in Mn- and Cr-doped Bi$_2$Te$_3$ topological insulators with varied thickness and doping content. By tracing the chirality of the Hall loops, we find that the Mn-type anomalous Hall effect with clockwise chirality is strengthened by the reduction of film thickness, which is opposite to that of the Cr-type anomalous Hall effect with counterclockwise chirality. We provide a phenomenological model to explain the evolution of magnetic order and anomalous Hall effect chirality in magnetically doped topological insulators.

preprint2021arXiv

Bridge the Vision Gap from Field to Command: A Deep Learning Network Enhancing Illumination and Details

With the goal of tuning up the brightness, low-light image enhancement enjoys numerous applications, such as surveillance, remote sensing and computational photography. Images captured under low-light conditions often suffer from poor visibility and blur. Solely brightening the dark regions will inevitably amplify the blur, thus may lead to detail loss. In this paper, we propose a simple yet effective two-stream framework named NEID to tune up the brightness and enhance the details simultaneously without introducing many computational costs. Precisely, the proposed method consists of three parts: Light Enhancement (LE), Detail Refinement (DR) and Feature Fusing (FF) module, which can aggregate composite features oriented to multiple tasks based on channel attention mechanism. Extensive experiments conducted on several benchmark datasets demonstrate the efficacy of our method and its superiority over state-of-the-art methods.

preprint2021arXiv

Conformal Wavefunctions for Graviton Amplitudes

The extended-BMS algebra of asymptotically flat spacetime contains an SO(3,1) subgroup that acts by conformal transformations on the celestial sphere. It is of interest to study the representations of this subgroup associated with gravitons. To reduce the equation of motion to a Schrodinger-like equation it is necessary to impose a non-covariant gauge condition. Using these solutions, leading-order gauge invariant Weyl scalars are then computed and decomposed into families of unitary principal series representations. An invertible holographic mapping is constructed between these unitary principal series operators and massless spin-2 perturbations of flat spacetime.

preprint2021arXiv

Descriptors of intrinsic hydrodynamic thermal transport: screening a phonon database in a machine learning approach

Machine learning techniques are used to explore the intrinsic origins of the hydrodynamic thermal transport and to find new materials interesting for science and engineering. The hydrodynamic thermal transport is governed intrinsically by the hydrodynamic scale and the thermal conductivity. The correlations between these intrinsic properties and harmonic and anharmonic properties, and a large number of compositional (290) and structural (1224) descriptors of 131 crystal compound materials are obtained, revealing some of the key descriptors that determines the magnitude of the intrinsic hydrodynamic effects, most of them related with the phonon relaxation times. Then, a trained black-box model is applied to screen more than 5000 materials. The results identify materials with potential technological applications. Understanding the properties correlated to hydrodynamic thermal transport can help to find new thermoelectric materials and on the design of new materials to ease the heat dissipation in electronic devices.

preprint2021arXiv

Dirac Fermion Cloning, Moir$\bf{é}$ Flat Bands and Magic Lattice Constants in Epitaxial Monolayer Graphene

Tuning interactions between Dirac states in graphene has attracted enormous interest because it can modify the electronic spectrum of the two-dimensional material, enhance electron correlations, and give rise to novel condensed-matter phases such as superconductors, Mott insulators, Wigner crystals and quantum anomalous Hall insulators. Previous works predominantly focus on the flat band dispersion of coupled Dirac states from different graphene layers. In this work, we propose a new route to realizing flat band physics in monolayer graphene under a periodic modulation from substrates. We take gaphene/SiC heterostructure as a role model and demonstrate experimentally the substrate modulation leads to Dirac fermion cloning and consequently, the proximity of the two Dirac cones of monolayer graphene in momentum space. Our theoretical modeling captures the cloning mechanism of Dirac states and indicates that flat bands can emerge at certain magic lattice constants of substrate when the period of modulation becomes nearly commensurate with the $(\sqrt{3}\times\sqrt{3})R30^{\circ}$ supercell of graphene. The results show that the epitaxial monolayer graphene is a promising platform for exploring exotic many-body quantum phases arising from interactions between Dirac electrons.

preprint2021arXiv

Extreme Suppression of Antiferromagnetic Order and Critical Scaling in a Two-Dimensional Random Quantum Magnet

Sr$_2$CuTeO$_6$ is a square-lattice Néel antiferromagnet with superexchange between first-neighbor $S=1/2$ Cu spins mediated by plaquette centered Te ions. Substituting Te by W, the affected impurity plaquettes have predominantly second-neighbor interactions, thus causing local magnetic frustration. Here we report a study of Sr$_2$CuTe$_{1-x}$W$_x$O$_6$ using neutron diffraction and $μ$SR techniques, showing that the Néel order vanishes already at $x = 0.025 \pm 0.005$. We explain this extreme order suppression using a two-dimensional Heisenberg spin model, demonstrating that a W-type impurity induces a deformation of the order parameter that decays with distance as $1/r^2$ at temperature $T=0$. The associated logarithmic singularity leads to loss of order for any $x>0$. Order for small $x>0$ and $T>0$ is induced by weak interplane couplings. In the nonmagnetic phase of Sr$_2$CuTe$_{1-x}$W$_x$O$_6$, the $μ$SR relaxation rate exhibits quantum critical scaling with a large dynamic exponent, $z \approx 3$, consistent with a random-singlet state.

preprint2021arXiv

Gaussian Process-Based Model Predictive Control for Overtaking

This paper proposes a novel framework for addressing the challenge of autonomous overtaking and obstacle avoidance, which incorporates the overtaking path planning into Gaussian Process-based model predictive control (GPMPC). Compared with the conventional control strategies, this approach has two main advantages. Firstly, combining Gaussian Process (GP) regression with a nominal model allows for learning from model mismatch and unmodeled dynamics, which enhances a simple model and delivers significantly better results. Due to the approximation for propagating uncertainties, we can furthermore satisfy the constraints and thereby safety of the vehicle is ensured. Secondly, we convert the geometric relationship between the ego vehicle and other obstacle vehicles into the constraints. Without relying on a higherlevel path planner, this approach substantially reduces the computational burden. In addition, we transform the state constraints under the model predictive control (MPC) framework into a soft constraint and incorporate it as relaxed barrier function into the cost function, which makes the optimizer more efficient. Simulation results reveal the usefulness of the proposed approach.

preprint2021arXiv

Hybrid simulation of energetic particles interacting with magnetohydrodynamics using a slow manifold algorithm and GPU acceleration

The hybrid method combining particle-in-cell and magnetohydrodynamics can be used to study the interaction between energetic particles and global plasma modes. In this paper we introduce the M3D-C1-K code, which is developed based on the M3D-C1 finite element code solving the magnetohydrodynamics equations, with a newly developed kinetic module simulating energetic particles. The particle pushing is done using a new algorithm by applying the Boris pusher to the classical Pauli particles to simulate the slow-manifold of particle orbits, with long-term accuracy and fidelity. The particle pushing can be accelerated using GPUs with a significant speedup. The moments of the particles are calculated using the $δf$ method, and are coupled into the magnetohydrodynamics simulation through pressure or current coupling schemes. Several linear simulations of magnetohydrodynamics modes driven by energetic particles have been conducted using M3D-C1-K, including fishbone, toroidal Alfvén eigenmodes and reversed shear Alfvén eigenmodes. Good agreement with previous results from other eigenvalue, kinetic and hybrid codes has been achieved.

preprint2021arXiv

Learning Invariant Representations across Domains and Tasks

Being expensive and time-consuming to collect massive COVID-19 image samples to train deep classification models, transfer learning is a promising approach by transferring knowledge from the abundant typical pneumonia datasets for COVID-19 image classification. However, negative transfer may deteriorate the performance due to the feature distribution divergence between two datasets and task semantic difference in diagnosing pneumonia and COVID-19 that rely on different characteristics. It is even more challenging when the target dataset has no labels available, i.e., unsupervised task transfer learning. In this paper, we propose a novel Task Adaptation Network (TAN) to solve this unsupervised task transfer problem. In addition to learning transferable features via domain-adversarial training, we propose a novel task semantic adaptor that uses the learning-to-learn strategy to adapt the task semantics. Experiments on three public COVID-19 datasets demonstrate that our proposed method achieves superior performance. Especially on COVID-DA dataset, TAN significantly increases the recall and F1 score by 5.0% and 7.8% compared to recently strong baselines. Moreover, we show that TAN also achieves superior performance on several public domain adaptation benchmarks.

preprint2021arXiv

Non-invasive Self-attention for Side Information Fusion in Sequential Recommendation

Sequential recommender systems aim to model users' evolving interests from their historical behaviors, and hence make customized time-relevant recommendations. Compared with traditional models, deep learning approaches such as CNN and RNN have achieved remarkable advancements in recommendation tasks. Recently, the BERT framework also emerges as a promising method, benefited from its self-attention mechanism in processing sequential data. However, one limitation of the original BERT framework is that it only considers one input source of the natural language tokens. It is still an open question to leverage various types of information under the BERT framework. Nonetheless, it is intuitively appealing to utilize other side information, such as item category or tag, for more comprehensive depictions and better recommendations. In our pilot experiments, we found naive approaches, which directly fuse types of side information into the item embeddings, usually bring very little or even negative effects. Therefore, in this paper, we propose the NOninVasive self-attention mechanism (NOVA) to leverage side information effectively under the BERT framework. NOVA makes use of side information to generate better attention distribution, rather than directly altering the item embedding, which may cause information overwhelming. We validate the NOVA-BERT model on both public and commercial datasets, and our method can stably outperform the state-of-the-art models with negligible computational overheads.

preprint2021arXiv

Recreation of the Periodic Table with an Unsupervised Machine Learning Algorithm

In 1869, the first draft of the periodic table was published by Russian chemist Dmitri Mendeleev. In terms of data science, his achievement can be viewed as a successful example of feature embedding based on human cognition: chemical properties of all known elements at that time were compressed onto the two-dimensional grid system for tabular display. In this study, we seek to answer the question of whether machine learning can reproduce or recreate the periodic table by using observed physicochemical properties of the elements. To achieve this goal, we developed a periodic table generator (PTG). The PTG is an unsupervised machine learning algorithm based on the generative topographic mapping (GTM), which can automate the translation of high-dimensional data into a tabular form with varying layouts on-demand. The PTG autonomously produced various arrangements of chemical symbols, which organized a two-dimensional array such as Mendeleev's periodic table or three-dimensional spiral table according to the underlying periodicity in the given data. We further showed what the PTG learned from the element data and how the element features, such as melting point and electronegativity, are compressed to the lower-dimensional latent spaces.

preprint2021arXiv

Shed Various Lights on a Low-Light Image: Multi-Level Enhancement Guided by Arbitrary References

It is suggested that low-light image enhancement realizes one-to-many mapping since we have different definitions of NORMAL-light given application scenarios or users' aesthetic. However, most existing methods ignore subjectivity of the task, and simply produce one result with fixed brightness. This paper proposes a neural network for multi-level low-light image enhancement, which is user-friendly to meet various requirements by selecting different images as brightness reference. Inspired by style transfer, our method decomposes an image into two low-coupling feature components in the latent space, which allows the concatenation feasibility of the content components from low-light images and the luminance components from reference images. In such a way, the network learns to extract scene-invariant and brightness-specific information from a set of image pairs instead of learning brightness differences. Moreover, information except for the brightness is preserved to the greatest extent to alleviate color distortion. Extensive results show strong capacity and superiority of our network against existing methods.

preprint2021arXiv

Towards Enhancing Fine-grained Details for Image Matting

In recent years, deep natural image matting has been rapidly evolved by extracting high-level contextual features into the model. However, most current methods still have difficulties with handling tiny details, like hairs or furs. In this paper, we argue that recovering these microscopic details relies on low-level but high-definition texture features. However, {these features are downsampled in a very early stage in current encoder-decoder-based models, resulting in the loss of microscopic details}. To address this issue, we design a deep image matting model {to enhance fine-grained details. Our model consists of} two parallel paths: a conventional encoder-decoder Semantic Path and an independent downsampling-free Textural Compensate Path (TCP). The TCP is proposed to extract fine-grained details such as lines and edges in the original image size, which greatly enhances the fineness of prediction. Meanwhile, to leverage the benefits of high-level context, we propose a feature fusion unit(FFU) to fuse multi-scale features from the semantic path and inject them into the TCP. In addition, we have observed that poorly annotated trimaps severely affect the performance of the model. Thus we further propose a novel term in loss function and a trimap generation method to improve our model's robustness to the trimaps. The experiments show that our method outperforms previous start-of-the-art methods on the Composition-1k dataset.

preprint2021arXiv

Unified gas-kinetic wave-particle methods VI: Disperse dilute gas-particle multiphase flow

In this paper, a unified gas-kinetic wave-particle scheme (UGKWP) for the disperse dilute gas-particle multiphase flow is proposed. The gas phase is always in the hydrodynamic regime. However, the particle phase covers different flow regimes from particle trajectory crossing to the hydrodynamic wave interaction with the variation of local particle phase Knudsen number. The UGKWP is an appropriate method for the capturing of the multiscale transport mechanism in the particle phase through its coupled wave-particle formulation. In the regime with intensive particle collision, the evolution of solid particle will be followed by the analytic wave with quasi-equilibrium distribution; while in the rarefied regime the non-equilibrium particle phase will be captured through particle tracking and collision, which plays a decisive role in recovering particle trajectory crossing behavior. The gas-kinetic scheme (GKS) is employed for the simulation of gas flow. In the highly collision regime for the particles, no particles will be sampled in UGKWP and the wave formulation for solid particle with the hydrodynamic gas phase will reduce the system to the two-fluid Eulerian model. On the other hand, in the collisionless regime for the solid particle, the free transport of solid particle will be followed in UGKWP, and coupled system will return to the Eulerian-Lagrangian formulation for the gas and particle. The scheme will be tested for in all flow regimes, which include the non-equilibrium particle trajectory crossing, the particle concentration under different Knudsen number, and the dispersion of particle flow with the variation of Stokes number. A experiment of shock-induced particle bed fluidization is simulated and the results are compared with experimental measurements. These numerical solutions validate suitability of the proposed scheme for the simulation of gas-particle multiphase flow.

Chang Liu

What is connected

Connect this record

See the researcher in context

Building this map preview

277 published item(s)

3D Dynamic Radio Map Prediction Using Vision Transformers for Low-Altitude Wireless Networks

Beyond Factor Aggregation: Gauge-Aware Low-Rank Server Representations for Federated LoRA

Biosignal Fingerprinting: A Cross-Modal PPG-ECG Foundation Model

Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment

Bridging Distance and Spectral Positional Encodings via Anchor-Based Diffusion Geometry Approximation

Chain-based Distillation for Effective Initialization of Variable-Sized Small Language Models

CoINS: Counterfactual Interactive Navigation via Skill-Aware VLM

Count Anything at Any Granularity

DCGL: Dual-Channel Graph Learning with Large Language Models for Knowledge-Aware Recommendation

Diffusion Model-Enhanced Environment Reconstruction in ISAC

Dynamic Water-Wave Tweezers

EgoIntrospect: An Egocentric Dataset and Benchmark for User-Centric Internal State Reasoning

FinDeepForecast: A Live Multi-Agent System for Benchmarking Deep Research Agents in Financial Forecasting

FinDeepResearch: Evaluating Deep Research Agents in Rigorous Financial Analysis

FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics

FreeText: Training-Free Text Rendering in Diffusion Transformers via Attention Localization and Spectral Glyph Injection

Generative structure search for efficient and diverse discovery of molecular and crystal structures

GREx: Generalized Referring Expression Segmentation, Comprehension, and Generation

Improving Human Image Animation via Semantic Representation Alignment

Learngene Search Across Multiple Datasets for Building Variable-Sized Models

Mamoda2.5: Enhancing Unified Multimodal Model with DiT-MoE

Masked Next-Scale Prediction for Self-supervised Scene Text Recognition

Motivating Effort with Information about Future Rewards

Report of the 5th PVUW Challenge: Towards More Diverse Modalities in Pixel-Level Understanding

SAM3-DMS: Decoupled Memory Selection for Multi-target Video Segmentation of SAM3

SEED: Targeted Data Selection by Weighted Independent Set

Self-organized MT Direction Maps Emerge from Spatiotemporal Contrastive Optimization

Structure preservation and emergent dissipation in stochastic wave equations with transport noise

Understanding Cross-Language Transfer Improvements in Low-Resource HTR: The Role of Sequence Modeling

Understanding Performance Collapse in Layer-Pruned Large Language Models via Decision Representation Transitions

Video Generation with Predictive Latents

ViewSAM: Learning View-aware Cross-modal Semantics for Weakly Supervised Cross-view Referring Multi-Object Tracking

VLingNav: Embodied Navigation with Adaptive Reasoning and Visual-Assisted Linguistic Memory

WaveFormer: Frequency-Time Decoupled Vision Modeling with Wave Equation

XPERT: Expert Knowledge Transfer for Effective Training of Language Models

AINav: Large Language Model-Based Adaptive Interactive Navigation

Influence of ambient temperature on cavitation bubble dynamics

MiMo-Audio: Audio Language Models are Few-Shot Learners

MTSP-LDP: A Framework for Multi-Task Streaming Data Publication under Local Differential Privacy

ReSPIRe: Informative and Reusable Belief Tree Search for Robot Probabilistic Search and Tracking in Unknown Environments

Tailoring Neel orders in Layered Topological Antiferromagnets

Foundation Model for Endoscopy Video Analysis via Large-scale Self-supervised Pre-train

LaCon: Late-Constraint Diffusion for Steerable Guided Image Synthesis

Benchmarking Graphormer on Large-Scale Molecular Modeling Datasets

Unified Gas-Kinetic Particle Method for Frequency-dependent Radiation Transport

6GAN: IPv6 Multi-Pattern Target Generation via Generative Adversarial Nets with Reinforcement Learning

A Comprehensive Study of Accelerating IPv6 Deployment

A Survey for Solving Mixed Integer Programming via Machine Learning

An efficient and easy-to-extend Matlab code of the Moving Morphable Component (MMC) method for three-dimensional topology optimization

An Empirical Study of Graphormer on Large-Scale Molecular Modeling Datasets

An Extendable Maneuver Management Framework with Fault-Tolerant Mechanism for Vehicle Platoon Control System in Highway Scenario

Atom-Specific Probing of Electron Dynamics in an Atomic Adsorbate by Time-Resolved X-ray Spectroscopy

Bayesian Sequential Stacking Algorithm for Concurrently Designing Molecules and Synthetic Reaction Networks

Constrained Deep One-Class Feature Learning For Classifying Imbalanced Medical Images

Crystal structure prediction with machine learning-based element substitution

Direct Molecular Conformation Generation

Disjoint Masking with Joint Distillation for Efficient Masked Image Modeling

DMF-Net: Dual-Branch Multi-Scale Feature Fusion Network for copy forgery identification of anti-counterfeiting QR code

Electromagnetic Follow-up Observations of Binary Neutron Star Mergers with Early Warnings from Decihertz Gravitational-wave Observatories

Extending the Fisher Information Matrix in Gravitational-wave Data Analysis

Generalizing to Unseen Domains: A Survey on Domain Generalization

Hot-tail electrons' impact on assimilation and injection penetration of D2 Shattered Pellet Injections

Improving the Robustness of Adversarial Attacks Using an Affine-Invariant Gradient Estimator

Increased Phase Coherence Length in a Porous Topological Insulator

Influences of the dissipative topological edge state on quantized transport in MnBi2Te4

Infrared Radiation of Graphene Electrothermal Film Triggered Alpha and Theta Brainwaves

Instance-Specific Feature Propagation for Referring Segmentation

Learning-based Predictive Beamforming for Integrated Sensing and Communication in Vehicular Networks

MAP-Gen: An Automated 3D-Box Annotation Flow with Multimodal Attention Point Generator

Medical Knowledge-Guided Deep Learning for Imbalanced Medical Image Classification

Multimodal Transformer for Automatic 3D Annotation and Object Detection

Nonlinear inhomogeneous Fokker-Planck models: energetic-variational structures and long time behavior

NTIRE 2022 Challenge on High Dynamic Range Imaging: Methods and Results

On the $α$-spectral radius of the $k$-uniform supertrees