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

Xiaohan Zhang

Xiaohan Zhang contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Machine LearningRoboticsComputer Visioncond-mat.mtrl-sciArtificial IntelligenceComputational Engineering, Finance, and ScienceDistributed, Parallel, and Cluster Computingeess.SPNeurons and Cognitionphysics.comp-phquant-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
14works
0followers
11topics
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

14 published item(s)

preprint2026arXiv

GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

We present GLM-4.1V-Thinking, GLM-4.5V, and GLM-4.6V, a family of vision-language models (VLMs) designed to advance general-purpose multimodal understanding and reasoning. In this report, we share our key findings in the development of the reasoning-centric training framework. We first develop a capable vision foundation model with significant potential through large-scale pre-training, which arguably sets the upper bound for the final performance. We then propose Reinforcement Learning with Curriculum Sampling (RLCS) to unlock the full potential of the model, leading to comprehensive capability enhancement across a diverse range of tasks, including STEM problem solving, video understanding, content recognition, coding, grounding, GUI-based agents, and long document interpretation. In a comprehensive evaluation across 42 public benchmarks, GLM-4.5V achieves state-of-the-art performance on nearly all tasks among open-source models of similar size, and demonstrates competitive or even superior results compared to closed-source models such as Gemini-2.5-Flash on challenging tasks including Coding and GUI Agents. Meanwhile, the smaller GLM-4.1V-9B-Thinking remains highly competitive-achieving superior results to the much larger Qwen2.5-VL-72B on 29 benchmarks. We open-source both GLM-4.1V-9B-Thinking and GLM-4.5V. We further introduce the GLM-4.6V series, open-source multimodal models with native tool use and a 128K context window. A brief overview is available at https://z.ai/blog/glm-4.6v. Code, models and more information are released at https://github.com/zai-org/GLM-V.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

ProcVLM: Learning Procedure-Grounded Progress Rewards for Robotic Manipulation

Long-horizon robotic manipulation requires dense feedback that reflects how a task advances through its procedural stages, not merely whether the final outcome is successful. Existing reward models often rely on trajectory-level success labels or time-based interpolation, which can conflate elapsed time with true task progress and therefore fail to capture unfinished steps, stagnation, and failure states. We present ProcVLM, a progress-aware vision-language model that learns procedure-grounded progress as a dense reward signal for manipulation. Rather than deriving progress from terminal outcomes or temporal proxies, ProcVLM grounds progress estimation in procedural structure and intra-stage visual change, and further adopts a reasoning-before-estimation paradigm that infers the remaining atomic actions before estimating task progress. Specifically, we construct this supervision by synthesizing frame-level subtask-semantic annotations, assigning progress budgets according to subtask structure, and distributing each budget based on intra-subtask visual change. To train ProcVLM at scale, we build a standardized procedural supervision synthesis pipeline and construct ProcCorpus-60M from 30 embodied datasets with 60M annotated frames, from which we derive ProcVQA for procedure-aware pretraining, with progress estimation as the central task alongside action segmentation and future planning. Experiments on ProcVQA and reward-model benchmarks show that ProcVLM improves embodied procedural reasoning and yields more discriminative trajectory-internal progress estimates than representative baselines, supporting its use as a dense reward model for downstream reward-guided policy optimization. Project page: https://procvlm.github.io/

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

RobotDiffuse: Diffusion-Based Motion Planning for Redundant Manipulators with the ROP Obstacle Avoidance Dataset

Redundant manipulators, with their higher Degrees of Freedom (DoFs), offer enhanced kinematic performance and versatility, making them suitable for applications like manufacturing, surgical robotics, and human-robot collaboration. However, motion planning for these manipulators is challenging due to increased DoFs and complex, dynamic environments. While traditional motion planning algorithms struggle with high-dimensional spaces, deep learning-based methods often face instability and inefficiency in complex tasks. This paper introduces RobotDiffuse, a diffusion model-based approach for motion planning in redundant manipulators. By integrating physical constraints with a point cloud encoder and replacing the U-Net structure with an encoder-only transformer, RobotDiffuse improves the model's ability to capture temporal dependencies and generate smoother, more coherent motion plans. We validate the approach using a complex simulator and release a new dataset, Robot-obtalcles-panda (ROP), with 35M robot poses and 0.14M obstacle avoidance scenarios. The highest overall score obtained in the experiment demonstrates the effectiveness of RobotDiffuse and the promise of diffusion models for motion planning tasks. The dataset can be accessed at https://github.com/ACRoboT-buaa/RobotDiffuse.

0 citations0 reviewsNo code linkedOpen access
preprint2026arXiv

VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation

Visual generative models have achieved remarkable progress in synthesizing photorealistic images and videos, yet aligning their outputs with human preferences across critical dimensions remains a persistent challenge. Though reinforcement learning from human feedback offers promise for preference alignment, existing reward models for visual generation face limitations, including black-box scoring without interpretability and potentially resultant unexpected biases. We present VisionReward, a general framework for learning human visual preferences in both image and video generation. Specifically, we employ a hierarchical visual assessment framework to capture fine-grained human preferences, and leverages linear weighting to enable interpretable preference learning. Furthermore, we propose a multi-dimensional consistent strategy when using VisionReward as a reward model during preference optimization for visual generation. Experiments show that VisionReward can significantly outperform existing image and video reward models on both machine metrics and human evaluation. Notably, VisionReward surpasses VideoScore by 17.2% in preference prediction accuracy, and text-to-video models with VisionReward achieve a 31.6% higher pairwise win rate compared to the same models using VideoScore. All code and datasets are provided at https://github.com/THUDM/VisionReward.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Deep Class Incremental Learning from Decentralized Data

In this paper, we focus on a new and challenging decentralized machine learning paradigm in which there are continuous inflows of data to be addressed and the data are stored in multiple repositories. We initiate the study of data decentralized class-incremental learning (DCIL) by making the following contributions. Firstly, we formulate the DCIL problem and develop the experimental protocol. Secondly, we introduce a paradigm to create a basic decentralized counterpart of typical (centralized) class-incremental learning approaches, and as a result, establish a benchmark for the DCIL study. Thirdly, we further propose a Decentralized Composite knowledge Incremental Distillation framework (DCID) to transfer knowledge from historical models and multiple local sites to the general model continually. DCID consists of three main components namely local class-incremental learning, collaborated knowledge distillation among local models, and aggregated knowledge distillation from local models to the general one. We comprehensively investigate our DCID framework by using different implementations of the three components. Extensive experimental results demonstrate the effectiveness of our DCID framework. The codes of the baseline methods and the proposed DCIL will be released at https://github.com/zxxxxh/DCIL.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Learning to Ground Objects for Robot Task and Motion Planning

Task and motion planning (TAMP) algorithms have been developed to help robots plan behaviors in discrete and continuous spaces. Robots face complex real-world scenarios, where it is hardly possible to model all objects or their physical properties for robot planning (e.g., in kitchens or shopping centers). In this paper, we define a new object-centric TAMP problem, where the TAMP robot does not know object properties (e.g., size and weight of blocks). We then introduce Task-Motion Object-Centric planning ({\bf TMOC}), a grounded TAMP algorithm that learns to ground objects and their physical properties with a physics engine. TMOC is particularly useful for those tasks that involve dynamic complex robot-multi-object interactions that can hardly be modeled beforehand. We have demonstrated and evaluated TMOC in simulation and using a real robot. Results show that TMOC outperforms competitive baselines from the literature in cumulative utility.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Learning to Guide Human Attention on Mobile Telepresence Robots with 360 Vision

Mobile telepresence robots (MTRs) allow people to navigate and interact with a remote environment that is in a place other than the person's true location. Thanks to the recent advances in 360 degree vision, many MTRs are now equipped with an all-degree visual perception capability. However, people's visual field horizontally spans only about 120 degree of the visual field captured by the robot. To bridge this observability gap toward human-MTR shared autonomy, we have developed a framework, called GHAL360, to enable the MTR to learn a goal-oriented policy from reinforcements for guiding human attention using visual indicators. Three telepresence environments were constructed using datasets that are extracted from Matterport3D and collected from a real robot respectively. Experimental results show that GHAL360 outperformed the baselines from the literature in the efficiency of a human-MTR team completing target search tasks.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Visually Grounded Task and Motion Planning for Mobile Manipulation

Task and motion planning (TAMP) algorithms aim to help robots achieve task-level goals, while maintaining motion-level feasibility. This paper focuses on TAMP domains that involve robot behaviors that take extended periods of time (e.g., long-distance navigation). In this paper, we develop a visual grounding approach to help robots probabilistically evaluate action feasibility, and introduce a TAMP algorithm, called GROP, that optimizes both feasibility and efficiency. We have collected a dataset that includes 96,000 simulated trials of a robot conducting mobile manipulation tasks, and then used the dataset to learn to ground symbolic spatial relationships for action feasibility evaluation. Compared with competitive TAMP baselines, GROP exhibited a higher task-completion rate while maintaining lower or comparable action costs. In addition to these extensive experiments in simulation, GROP is fully implemented and tested on a real robot system.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Calibration-Free Vector Magnetometry Using Nitrogen-Vacancy Center in Diamond Integrated with Optical Vortex Beam

We report a new method to determine the orientation of individual nitrogen-vacancy (NV) centers in a bulk diamond and use them to realize a calibration-free vector magnetometer with nanoscale resolution. Optical vortex beam is used for optical excitation and scanning the NV center in a [111]-oriented diamond. The scanning fluorescence patterns of NV center with different orientations are completely different. Thus, the orientation information on each NV center in the lattice can be known directly without any calibration process. Further, we use three differently oriented NV centers to form a magnetometer and reconstruct the complete vector information on the magnetic field based on the optically detected magnetic resonance(ODMR) technique. Compared with previous schemes to realize vector magnetometry using an NV center, our method is much more efficient and is easily applied in other NV-based quantum sensing applications.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Episodic memory governs choices: An RNN-based reinforcement learning model for decision-making task

Typical methods to study cognitive function are to record the electrical activities of animal neurons during the training of animals performing behavioral tasks. A key problem is that they fail to record all the relevant neurons in the animal brain. To alleviate this problem, we develop an RNN-based Actor-Critic framework, which is trained through reinforcement learning (RL) to solve two tasks analogous to the monkeys' decision-making tasks. The trained model is capable of reproducing some features of neural activities recorded from animal brain, or some behavior properties exhibited in animal experiments, suggesting that it can serve as a computational platform to explore other cognitive functions. Furthermore, we conduct behavioral experiments on our framework, trying to explore an open question in neuroscience: which episodic memory in the hippocampus should be selected to ultimately govern future decisions. We find that the retrieval of salient events sampled from episodic memories can effectively shorten deliberation time than common events in the decision-making process. The results indicate that salient events stored in the hippocampus could be prioritized to propagate reward information, and thus allow decision-makers to learn a strategy faster.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Deep Reinforcement Learning for Fog Computing-based Vehicular System with Multi-operator Support

This paper studies the potential performance improvement that can be achieved by enabling multi-operator wireless connectivity for cloud/fog computing-connected vehicular systems. Mobile network operator (MNO) selection and switching problem is formulated by jointly considering switching cost, quality-of-service (QoS) variations between MNOs, and the different prices that can be charged by different MNOs as well as cloud and fog servers. A double deep Q network (DQN) based switching policy is proposed and proved to be able to minimize the long-term average cost of each vehicle with guaranteed latency and reliability performance. The performance of the proposed approach is evaluated using the dataset collected in a commercially available city-wide LTE network. Simulation results show that our proposed policy can significantly reduce the cost paid by each fog/cloud-connected vehicle with guaranteed latency services.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Finite Element Approximation of Finite Deformation Dislocation Mechanics

We develop and demonstrate the first general computational tool for finite deformation static and dynamic dislocation mechanics. A finite element formulation of finite deformation (Mesoscale) Field Dislocation Mechanics theory is presented. The model is a minimal enhancement of classical crystal/$J_2$ plasticity that fundamentally accounts for polar/excess dislocations at the mesoscale. It has the ability to compute the static and dynamic finite deformation stress fields of arbitrary (evolving) dislocation distributions in finite bodies of arbitrary shape and elastic anisotropy under general boundary conditions. This capability is used to present a comparison of the static stress fields, at finite and small deformations, for screw and edge dislocations, revealing heretofore unexpected differences. The computational framework is verified against the sharply contrasting predictions of geometrically linear and nonlinear theories for the stress field of a spatially homogeneous dislocation distribution in the body, as well as against other exact results of the theory. Verification tests of the time-dependent numerics are also presented. Size effects in crystal and isotropic versions of the theory are shown to be a natural consequence of the model and are validated against available experimental data. With inertial effects incorporated, the development of an (asymmetric) propagating Mach cone is demonstrated in the finite deformation theory when a dislocation moves at speeds greater than the linear elastic shear wave speed of the material.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Plasticity without phenomenology: a first step

A novel, concurrent multiscale approach to meso/macroscale plasticity is demonstrated. It utilizes a carefully designed coupling of a partial differential equation (pde) based theory of dislocation mediated crystal plasticity with time-averaged inputs from microscopic Dislocation Dynamics (DD), adapting a state-of-the-art mathematical coarse-graining scheme. The stress-strain response of mesoscopic samples at realistic, slow, loading rates up to appreciable values of strain is obtained, with significant speed-up in compute time compared to conventional DD. Effects of crystal orientation, loading rate, and the ratio of the initial mobile to sessile dislocation density on the macroscopic response, for both load and displacement controlled simulations are demonstrated. These results are obtained without using any phenomenological constitutive assumption, except for thermal activation which is not a part of microscopic DD. The results also demonstrate the effect of the internal stresses on the collective behavior of dislocations, manifesting, in a set of examples, as a Stage I to Stage II hardening transition.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Scale Calibrated Training: Improving Generalization of Deep Networks via Scale-Specific Normalization

Standard convolutional neural networks(CNNs) require consistent image resolutions in both training and testing phase. However, in practice, testing with smaller image sizes is necessary for fast inference. We show that trivially evaluating low-resolution images on networks trained with high-resolution images results in a catastrophic accuracy drop in standard CNN architectures. We propose a novel training regime called Scale calibrated Training(SCT) which allows networks to learn from various scales of input simultaneously. By taking advantages of SCT, single network can provide decent accuracy at test time in response to multiple test scales. In our analysis, we surprisingly find that vanilla batch normalization can lead to sub-optimal performance in SCT. Therefore, a novel normalization scheme called Scale-Specific Batch Normalization is equipped to SCT in replacement of batch normalization. Experiment results show that SCT improves accuracy of single Resnet-50 on ImageNet by 1.7% and 11.5% accuracy when testing on image sizes of 224 and 128 respectively.

0 citations0 reviewsNo code linkedOpen access