BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Yanyan Zhao

Yanyan Zhao contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Artificial IntelligenceComputation and Languagecond-mat.mtrl-sciCryptography and SecurityInformation Retrievalphysics.chem-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
8works
0followers
6topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

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

Published work

8 published item(s)

preprint2026arXiv

Learning to Learn from Multimodal Experience

Experience-driven learning has emerged as a promising paradigm for enabling agents to improve from interaction trajectories by accumulating and reusing past experience. However, existing approaches are predominantly developed in textual settings and rely on manually designed memory schemas, limiting their applicability to multimodal environments. In real-world scenarios, experience is inherently multimodal, involving heterogeneous signals across perception, reasoning, and action, which makes effective memory design significantly more challenging. In particular, the optimal way to structure and utilize multimodal experience is highly task-dependent and evolves over time, rendering fixed memory designs insufficient. In this work, we propose a new paradigm, learning to learn from multimodal experience, which shifts memory design from a predefined component to an adaptive and learnable process. Our framework enables agents to dynamically construct, organize, and utilize memory based on task requirements and interaction history, effectively learning how to structure experience for improved performance. Experiments demonstrate that adaptive memory design substantially enhances agent performance and generalization across multimodal tasks, highlighting the critical role of learning memory mechanisms in experience-driven learning.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Rethinking Experience Utilization in Self-Evolving Language Model Agents

Self-evolving agents improve by accumulating and reusing experience from past interactions. Existing work has largely focused on how experience is constructed, represented, and updated, while paying less attention to how experience should be used during runtime decision-making. As a result, most agents rely on rigid usage strategies, either injecting experience once at initialization or at every step, without considering whether it is needed for the current decision. This paper studies experience utilization as a critical design dimension of self-evolving agents. We ask whether agents benefit from interweaving experience use with decision-making, so that experience is invoked only when additional guidance is needed. To examine this question, we introduce {ExpWeaver}, a lightweight instantiation that leaves experience construction unchanged and modifies only runtime utilization by exposing experience as an optional resource during reasoning. Across four representative frameworks, seven LLM backbones, and three types of environments, ExpWeaver consistently achieves the best performance among different utilization strategies. Reinforcement learning experiments further show that this behavior can be amplified through training. Usage-pattern, causal ablation, and entropy-based analyses reveal that ExpWeaver enables agents to invoke experience selectively, at beneficial decision points, and under higher reasoning uncertainty. Overall, our findings call for a shift from merely studying \emph{what} experience to store toward understanding \emph{how} and \emph{when} experience should enter decision-making.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

Safety Geometry Collapse in Multimodal LLMs and Adaptive Drift Correction

Multimodal large language models (MLLMs) often fail to transfer safety capabilities learned in the text modality to semantically equivalent non-text inputs, revealing a persistent multimodal safety gap. We study this gap from a representation-geometric perspective by analyzing a text-aligned refusal direction and a modality-induced drift direction. We show that multimodal inputs compress the usable separation along the refusal direction, making it no longer reliable for identifying and refusing harmful inputs. We refer to this failure mode as Safety Geometry Collapse. We quantify it through conditional refusal separability and show that stronger modality-induced drift is consistently associated with weaker refusal separability and higher attack success rates. We then validate the causal role of modality-induced drift through a fixed-strength activation intervention: counteracting the estimated drift restores refusal separability and improves multimodal safety. After drift correction, we further observe self-rectification, where the model recovers its ability to recognize and refuse harmful multimodal inputs during forward dynamics. This effect also provides an internal signal of the model's perceived harmfulness of each input. Motivated by this signal, we propose ReGap, a training-free inference-time method that adaptively corrects modality drift using self-rectification. Experiments across multiple multimodal safety benchmarks and utility benchmarks demonstrate the effectiveness of ReGap, which significantly improves the safety of MLLMs without compromising general capabilities. Our findings highlight representation-level modality alignment as a crucial direction for real-time safety improvement and for building safer, more reliable MLLMs.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules

Defending against jailbreak attacks is crucial for the safe deployment of Large Language Models (LLMs). Recent research has attempted to improve safety by training models to reason over safety rules before responding. However, a key issue lies in determining what form of safety reasoning effectively defends against jailbreak attacks, which is difficult to explicitly design or directly obtain. To address this, we propose \textbf{STAR-S} (\textbf{S}elf-\textbf{TA}ught \textbf{R}easoning based on \textbf{S}afety rules), a framework that integrates the learning of safety rule reasoning into a self-taught loop. The core of STAR-S involves eliciting reasoning and reflection guided by safety rules, then leveraging fine-tuning to enhance safety reasoning. Repeating this process creates a synergistic cycle. Improvements in the model's reasoning and interpretation of safety rules allow it to produce better reasoning data under safety rule prompts, which is then utilized for further training. Experiments show that STAR-S effectively defends against jailbreak attacks, outperforming baselines. Code is available at: https://github.com/pikepokenew/STAR_S.git.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

MACSA: A Multimodal Aspect-Category Sentiment Analysis Dataset with Multimodal Fine-grained Aligned Annotations

Multimodal fine-grained sentiment analysis has recently attracted increasing attention due to its broad applications. However, the existing multimodal fine-grained sentiment datasets most focus on annotating the fine-grained elements in text but ignore those in images, which leads to the fine-grained elements in visual content not receiving the full attention they deserve. In this paper, we propose a new dataset, the Multimodal Aspect-Category Sentiment Analysis (MACSA) dataset, which contains more than 21K text-image pairs. The dataset provides fine-grained annotations for both textual and visual content and firstly uses the aspect category as the pivot to align the fine-grained elements between the two modalities. Based on our dataset, we propose the Multimodal ACSA task and a multimodal graph-based aligned model (MGAM), which adopts a fine-grained cross-modal fusion method. Experimental results show that our method can facilitate the baseline comparison for future research on this corpus. We will make the dataset and code publicly available.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Exploring Periodicity and Interactivity in Multi-Interest Framework for Sequential Recommendation

Sequential recommendation systems alleviate the problem of information overload, and have attracted increasing attention in the literature. Most prior works usually obtain an overall representation based on the user's behavior sequence, which can not sufficiently reflect the multiple interests of the user. To this end, we propose a novel method called PIMI to mitigate this issue. PIMI can model the user's multi-interest representation effectively by considering both the periodicity and interactivity in the item sequence. Specifically, we design a periodicity-aware module to utilize the time interval information between user's behaviors. Meanwhile, an ingenious graph is proposed to enhance the interactivity between items in user's behavior sequence, which can capture both global and local item features. Finally, a multi-interest extraction module is applied to describe user's multiple interests based on the obtained item representation. Extensive experiments on two real-world datasets Amazon and Taobao show that PIMI outperforms state-of-the-art methods consistently.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Enhanced Ferromagnetism of CrI3 Bilayer by Self-Intercalation

Two-dimensional (2D) ferromagnets with high Curie temperature have long been the pursuit for electronic and spintronic applications. CrI3 is a rising star of intrinsic 2D ferromagnets, however, it suffers from weak exchange coupling. Here we propose a general strategy of self-intercalation to achieve enhanced ferromagnetism in bilayer CrI3. We showed that filling either Cr or I atoms into the van der Waals gap of stacked and twisted CrI3 bilayers can induce the double exchange effect and significantly strengthen the interlayer ferromagnetic coupling. According to our first-principles calculations, the intercalated native atoms act as covalent bridge between two CrI3 layers and lead to discrepant oxidation states for the Cr atoms. These theoretical results offer a facile route to achieve high-Curie-temperature 2D magnets for device implementation.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Selective C-C Coupling by Spatially Confined Dimeric Metal Centers

Direct conversion of carbon dioxide (CO2) to high-energy fuels and high-value chemicals is a fascinating sustainable strategy. For most of the current electrocatalysts for CO2 reduction, however, multi-carbon products are inhibited by large overpotentials and low selectivity. For practical applications, there remains a big gap of knowledge in proper manipulation of the C-C coupling process. Herein, we exploit dispersed 3d transition metal dimers as spatially confined dual reaction centers for selective reduction of CO2 to liquid fuels. Various nitrogenated holey carbon monolayers are shown to be promising templates to stabilize these metal dimers and dictate their electronic structures, allowing precise control of the catalytic activity and product selectivity. By comprehensive first-principles calculations, we screen the suitable transition metal dimers that universally have high activity for ethanol (C2H5OH). Furthermore, remarkable selectivity for C2H5OH against other C1 and C2 products is found for Fe2 dimer anchored on C2N monolayer. The correlation between the activity and d band center of the supported metal dimer as well as the role of electronic coupling between the metal dimer and the carbon substrates are thoroughly elucidated.

0 citations0 reviewsNo code linkedOpen access