Showing 1–50 of 171 results for author: Zhai, W

Reaction-Transformation-Aware Flow Matching for Generalizable Transition State Generation

Authors: Kaipeng Zeng, Wenxi Zhai, Shengrui Xu, Jie Zhao, Bowen Li, Shiyue Wang, Junchi Yan, Tong Zhu

Abstract: Transition-state (TS) structures define the energetic barriers and mechanistic pathways of elementary chemical reactions, yet their identification remains computationally demanding because conventional saddle-point searches require expensive quantum-mechanical calculations. Recent machine-learning approaches have accelerated TS generation by predicting structures from reaction endpoint information… ▽ More Transition-state (TS) structures define the energetic barriers and mechanistic pathways of elementary chemical reactions, yet their identification remains computationally demanding because conventional saddle-point searches require expensive quantum-mechanical calculations. Recent machine-learning approaches have accelerated TS generation by predicting structures from reaction endpoint information, but they primarily learn geometric correspondence between endpoints and TSs, leaving the structural transformations underlying elementary reactions implicitly represented. To address this limitation, we introduce TransTS, a reaction-transformation-aware framework for generalizable TS generation from atom-mapped reactant-product pairs. TransTS explicitly learns atom-level structural transformations between reaction endpoints and integrates them with a unified atom-aligned geometric representation of reactants, TSs and products, enabling reaction-aware equivariant generation of TS geometries. TransTS is designed to provide reliable TS initial guesses for subsequent quantum-chemical refinement, where generated structures are evaluated not only by geometric similarity but also by their ability to converge to validated saddle points and recover the intended reaction pathways. Across IID and zero-shot OOD benchmarks, TransTS demonstrates improved TS initialization quality, with particularly strong generalization to unseen reaction distributions. On the challenging GDB-10-rxn and GDB-17-rxn OOD benchmarks, TransTS generates TS candidates that more frequently converge to validated saddle points and recover the intended elementary reactions after refinement than existing approaches under the same training regime. Scaling reaction coverage and model capacity further improves both geometric fidelity and refinement outcomes. △ Less

Submitted 14 August, 2026; originally announced August 2026.

History-informed Lagrangian Neural Networks

Authors: Tianshuo Zhang, Xianglei Xing, Wenzhe Zhai, Jia Gao, He Cao

Abstract: Forecasting the long-horizon evolution of mechanical systems from position-only observations is a pivotal yet difficult task, as hidden velocities and trajectory-specific physical properties must be inferred simultaneously. Although physics-guided neural networks like Lagrangian Neural Networks (LNNs) guarantee physical plausibility, they generally require complete state inputs and lack adaptabili… ▽ More Forecasting the long-horizon evolution of mechanical systems from position-only observations is a pivotal yet difficult task, as hidden velocities and trajectory-specific physical properties must be inferred simultaneously. Although physics-guided neural networks like Lagrangian Neural Networks (LNNs) guarantee physical plausibility, they generally require complete state inputs and lack adaptability to changing system parameters. To break these limitations, we introduce History-informed Lagrangian Neural Networks (HiLNN). Grounded in the insight that temporal position sequences implicitly encode underlying dynamics, HiLNN employs a recurrent encoder to extract a latent context from history. This context not only reconstructs the unobserved initial velocity but also adaptively modulates the mass matrix, potential energy, and damping coefficients of a structured Lagrangian system. By leveraging a differentiable RK4 rollout scheme, the entire pipeline is optimized end-to-end under multi-step trajectory supervision and energy-consistency regularization. Empirical evaluations across conservative, dissipative, and heterogeneous variable-parameter systems show that HiLNN delivers superior long-term prediction accuracy and maintains precise energy profiles compared to state-of-the-art baselines. The source code is publicly available at https://github.com/yingtian22/History-informed-LNN. △ Less

Submitted 13 August, 2026; originally announced August 2026.

Comments: 15 pages, 5 figures. Accepted to the 9th Chinese Conference on Pattern Recognition and Computer Vision (PRCV 2026) as an oral paper

MAJEPPA: Morphing and Assessing in a Unified Piano Performance Space

Authors: Jinwen Zhou, Huan Zhang, Weixi Zhai, Jinhua Liang, Aidan O. T. Hogg, Simon Dixon

Abstract: We present MAJEPPA, a self-supervised framework to learn piano performance representations that span the full skill spectrum, from beginner practice sessions to virtuoso concert recordings. We curate the MAJEPPA dataset, comprising ~4,000 annotated recordings across six expertise levels and six recording contexts. We adapt a single pre-trained MIDI autoregressive model with a joint objective: next… ▽ More We present MAJEPPA, a self-supervised framework to learn piano performance representations that span the full skill spectrum, from beginner practice sessions to virtuoso concert recordings. We curate the MAJEPPA dataset, comprising ~4,000 annotated recordings across six expertise levels and six recording contexts. We adapt a single pre-trained MIDI autoregressive model with a joint objective: next-token prediction learns score-conditioned performance generation at various skill levels, while InfoNCE and supervised contrastive losses align abstract score and performance representations in a joint embedding space. The proposed model both generates and understands performances in a unified framework. By introducing the EVPMR benchmark, a suite of downstream tasks spanning quality assessment, competition ranking, mistake and technique classification, we evaluate the learnt representations, demonstrating progress towards a real-world model for the piano performance space. △ Less

Submitted 11 August, 2026; originally announced August 2026.

LogiShot: Logically Coherent Cross-Shot Video Generation

Authors: Shuai Guo, Yuhang Yang, Zeyu Zhang, Pengfei Yu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Generating cross-shot videos that are logically connected is essential for content creation. Currently, most cross-shot video-generation workflows, such as short-drama production, still rely on isolated textual scripts or explicit reference images to specify the generated content. Consequently, when user instructions are underspecified or ambiguous, a generated clip may appear visually plausible o… ▽ More Generating cross-shot videos that are logically connected is essential for content creation. Currently, most cross-shot video-generation workflows, such as short-drama production, still rely on isolated textual scripts or explicit reference images to specify the generated content. Consequently, when user instructions are underspecified or ambiguous, a generated clip may appear visually plausible on its own but fail to align with the overall narrative, leading to disjointed content. We argue that achieving cross-shot logical coherence in video generation requires establishing logical connections across shots and maintaining visual consistency. To this end, we propose LogiShot, which incorporates information through two complementary paths: 1) LogiShot jointly encodes the context video and other conditioning signals, yielding dense multimodal cues that provide visual-semantic evidence for cross-shot generation; 2) the model maintains a visual memory of the context video throughout generation to preserve visual consistency across shots. Additionally, we construct a dataset with 110K samples and a dedicated benchmark for evaluating cross-shot logical coherence. Experiments demonstrate that LogiShot consistently outperforms existing baselines in terms of logical coherence across multiple shots. Model and data will be made publicly available. △ Less

Submitted 9 August, 2026; originally announced August 2026.

Science Edge Evaluation: SEE the Missing Step Toward Real Scientific Discovery

Authors: Taolin Han, Yuchen Zhang, Jinghang Wang, Yun Wu, Wai Yuet Chiu, Zhaohai Li, Yifei Zhang, Jinxin Wang, Yuhao Zhou, Chen Zhao, Jiajia Li, Jiaxin Li, Qile Jin, Kewei Sun, Shuang Wu, Weiqi Zhai, Renquan Lv, Junchao Li, Ruodan Chen, Qingteng Chen, Zhibo Yang, Hu Wei, Lin Qu, Shuai Bai, Bing Zhao

Abstract: Large language models (LLMs) are increasingly involved in scientific discovery, yet it remains unclear whether they can support complex real laboratory science. Here we introduce Science Edge Evaluation (SEE), a multimodal benchmark of expert-curated questions grounded in peer-reviewed literature and experimental practice in chemistry, biology, and materials science. Evaluation of 19 multimodal la… ▽ More Large language models (LLMs) are increasingly involved in scientific discovery, yet it remains unclear whether they can support complex real laboratory science. Here we introduce Science Edge Evaluation (SEE), a multimodal benchmark of expert-curated questions grounded in peer-reviewed literature and experimental practice in chemistry, biology, and materials science. Evaluation of 19 multimodal large language models (MLLMs) shows that even the best-performing model reaches only 48.7% accuracy. Moreover, general-purpose models outperform science-specialized models on average. In the visual-agent evaluation, the use of tools increases the best accuracy to 52.7%. Tool use can expand the information available to models, but more information does not necessarily lead to reliable scientific reasoning. The key challenge is whether models can manage tool-derived information within the boundaries of the original experimental evidence. Together, these findings reveal that current MLLMs still cannot reliably make justified and evidence-bounded inferences from experimental results, which is an essential capability in real scientific discovery. Bridging this gap requires MLLMs to transition from explaining established scientific concepts to deriving novel and evidence-based insights from experimental data. △ Less

Submitted 7 August, 2026; originally announced August 2026.

Right Answer, Wrong Method: Shortcut Hacking Misleads the Evaluation of LLM Reasoning on Frontier Science Benchmarks

Authors: Xuan Ren, Weiqi Zhai, Tianle Pu, Yihua Zhu, Yihua Zhu, Hu Wei, Bing Zhao

Abstract: Scientific reasoning benchmarks typically evaluate large language models (LLMs) using final-answer accuracy. However, a correct answer does not necessarily demonstrate the reasoning capability targeted by the problem. We identify Solution Hacking, a failure mode in which an LLM reaches the correct answer through invalid shortcuts, such as numerical search, enumeration, guessing, or answer-first ve… ▽ More Scientific reasoning benchmarks typically evaluate large language models (LLMs) using final-answer accuracy. However, a correct answer does not necessarily demonstrate the reasoning capability targeted by the problem. We identify Solution Hacking, a failure mode in which an LLM reaches the correct answer through invalid shortcuts, such as numerical search, enumeration, guessing, or answer-first verification, without providing a valid task-targeted derivation. We systematically analyze this phenomenon across difficulty levels, scientific domains, and frontier models. Solution hacking increases sharply with benchmark difficulty, from 2.2\% on common problems to 28.3\% on Olympiad-level problems and 37.4\% on HLE. Moreover, 8.2\%-44.1\% of answers credited as correct across frontier models are identified as hacked solutions. We further develop expert-inspired anti-hacking strategies, including an automatic judge and a test-time instruction. The results show that suppressing shortcut behavior substantially reduces reported accuracy while having a smaller effect on correct and non-hacked accuracy. These findings reveal that answer-only evaluation can overestimate the scientific reasoning capabilities of frontier LLMs. △ Less

Submitted 3 August, 2026; originally announced August 2026.

Comments: working in progress

Before the Action: Benchmarking LLMs on Prospective Hypothesis Discovery

Authors: Tianyun Zhong, Wangyi Jiang, Wei Wang, Xuanang Chen, Yaojie Lu, Shiwei Ye, Yuzhen Shi, Boyu Yang, Jinghang Wang, Han Li, Weiqi Zhai, Bing Zhao, Hu Wei, Haiyang Yu, Yongbin Li, Hongyu Lin, Le Sun, Xianpei Han

Abstract: Large language models (LLMs) excel at answering pre-specified questions, yet their ability to navigate the open-ended, pre-conclusion stage of discovery remains largely unmeasured. We introduce Prospective Hypothesis Discovery (PHD), which asks models to autonomously construct grounded, discriminative, and testable hypothesis spaces from inconclusive evidence, including anomalous observations and… ▽ More Large language models (LLMs) excel at answering pre-specified questions, yet their ability to navigate the open-ended, pre-conclusion stage of discovery remains largely unmeasured. We introduce Prospective Hypothesis Discovery (PHD), which asks models to autonomously construct grounded, discriminative, and testable hypothesis spaces from inconclusive evidence, including anomalous observations and fragmented records, to guide subsequent investigation. To evaluate this capability, we introduce HypoArena, comprising HypoData, a benchmark of 988 cases across six scientific and analytical domains, and HypoEval, an evaluation framework for open-ended hypothesis sets. To construct HypoData at scale, we propose Retrospective Context Regression, a Forge--Audit pipeline that reconstructs pre-conclusion contexts from completed expert documents by removing explicit conclusions, target hypotheses, and retrospective causal attributions while preserving the factual substrate. Because PHD admits multiple valid outputs, HypoEval combines bidirectional pairwise judgments with Bradley--Terry--Davidson aggregation for ranking and six-dimensional rubric scoring for diagnosis. Experiments on 15 frontier LLMs reveal clear capability stratification and model-dependent effects of structured analytical skills, with gains for several lower-performing models on HypoArena but regressions for other systems, including a top-performing model. Compared with absolute rubric scoring, arena evaluation resolves finer-grained differences among models, with aggregated rankings showing strong agreement with human experts and an independent judge. Together, these results support treating PHD as a distinct target for evaluating how LLMs formulate investigative directions when final conclusions are withheld. Our code and data are publicly available at github.com/SKYLENAGE-AI/HypoArena and github.com/SKYLENAGE-AI/HypoArena. △ Less

Submitted 17 July, 2026; originally announced July 2026.

LongMedBench: Benchmarking Medical Agents for Long-Horizon Clinical Decision-Making

Authors: Zihan Xu, Yanzhen Chen, Xiaocheng Zhang, Zhiting Fan, Weiqi Zhai, Hongxia Xu, Zuozhu Liu

Abstract: In this work, we introduce LongMedBench, a real-world EHR-based benchmark for long-horizon clinical decision-making. Prior evaluations of LLM-based medical agents have largely emphasized short-context knowledge QA and tool use. However, real-world medical care is inherently longitudinal, and clinicians must aggregate evidence across repeated visits, tests, and evolving treatments. Therefore, long-… ▽ More In this work, we introduce LongMedBench, a real-world EHR-based benchmark for long-horizon clinical decision-making. Prior evaluations of LLM-based medical agents have largely emphasized short-context knowledge QA and tool use. However, real-world medical care is inherently longitudinal, and clinicians must aggregate evidence across repeated visits, tests, and evolving treatments. Therefore, long-horizon interaction is essential for realistic assessment. LongMedBench is constructed via a reproducible pipeline that integrates MIMIC-IV admission records and clinical notes into time-series event streams and long-context memory datasets, enabling long-horizon, multi-session interactions between agents and a clinical environment. It comprises 335 patients, with 19.72 inpatient visits per patient on average and 44.91 medical events per visit. Guided by the long-horizon decision process, we propose an evaluation taxonomy with three suites: fact-based QA, temporal reasoning, and long-horizon decision-making. This taxonomy measures how agents understand and leverage historical patient information over extended horizons. Our experiments show that while recent LLMs can make good use of explicit timestamps, they have challenges in implicit time inference; The RAG and agent memory system can improve the performance of information retrieval tasks, but the performance of decision-making tasks is highly dependent on the model's immediate context. △ Less

Submitted 13 July, 2026; v1 submitted 10 July, 2026; originally announced July 2026.

Comments: Submitted manuscript prior to peer review in MICCAI 2026

Movable Antennas for Robust Wireless Sensing via Joint Cramér-Rao Bound and Sidelobe Minimization

Authors: Wenyan Ma, Lipeng Zhu, Weitong Zhai, Rui Zhang

Abstract: This paper presents a novel design approach for movable antenna (MA)-enabled wireless sensing systems by jointly minimizing the Cramér-Rao bound (CRB) and the maximum sidelobe level (MSL) of the ambiguity function via antenna position optimization. In particular, the mean squared error (MSE) of angle-of-arrival (AoA) estimation is decomposed into a local estimation error within the mainlobe of the… ▽ More This paper presents a novel design approach for movable antenna (MA)-enabled wireless sensing systems by jointly minimizing the Cramér-Rao bound (CRB) and the maximum sidelobe level (MSL) of the ambiguity function via antenna position optimization. In particular, the mean squared error (MSE) of angle-of-arrival (AoA) estimation is decomposed into a local estimation error within the mainlobe of the ambiguity function (i.e., CRB) and an additional ambiguity error caused by its sidelobes. Since the MSE is dominated by the CRB in the high-signal-to-noise ratio (SNR) regime but by the sidelobes of the ambiguity function in the low-SNR regime, our analysis reveals a fundamental trade-off between CRB minimization and MSL minimization in the moderate-SNR regime. Specifically, minimizing the CRB prefers a narrower mainlobe, where antennas are concentrated near the two edges of the one-dimensional (1-D) movement region; whereas minimizing the MSL favors a wider mainlobe, where antennas are distributed more densely near the center of the movement region. Inspired by this and to ensure robust sensing performance across different SNR regimes, we formulate an optimization problem to minimize the CRB subject to a prescribed MSL constraint via antenna position optimization. An efficient successive convex approximation (SCA) algorithm is developed to optimize the antenna position vector (APV), and a 1-D linear search method is proposed to determine the optimal MSL threshold that minimizes the actual MSE for any given SNR. Numerical results demonstrate that the proposed scheme effectively balances the trade-off between MSL and CRB minimization, thus achieving a significantly lower AoA estimation MSE across the entire SNR range compared to conventional uniform and non-uniform fixed-position antenna (FPA) arrays. △ Less

Submitted 22 June, 2026; originally announced June 2026.

Tunable Flat Bands and magnetism in Triangulene-based Superatomic Graphene

Authors: Wenya Zhai, Tingfeng Zhang, Fengkun Chen, Xiuqin Lu, Yunlong Xia, Zengfu Ou, Ye Chen, Donghui Guo, Meifang Zhu, Zhengfei Wang, Jingcheng Li

Abstract: Superatomic graphene platforms host a rich portfolio of flat-band-driven exotic quantum properties, yet their experimental realization remains challenging. Here, we report the bottom-up on-surface synthesis of superatomic graphene using phosphorus-doped triangulene as building blocks. Scanning tunneling microscopy and spectroscopy measurements resolve the well-defined honeycomb lattice of as-fabri… ▽ More Superatomic graphene platforms host a rich portfolio of flat-band-driven exotic quantum properties, yet their experimental realization remains challenging. Here, we report the bottom-up on-surface synthesis of superatomic graphene using phosphorus-doped triangulene as building blocks. Scanning tunneling microscopy and spectroscopy measurements resolve the well-defined honeycomb lattice of as-fabricated superatomic graphene and demonstrate the characteristic Dirac band and flat band electronic structures. Density functional theory calculations reveal that the flat bands originate from the in-plane p$_x,_y$-like frontier orbitals of the phosphorus-doped triangulene units, leading to intrinsic half-metallic behavior. Furthermore, oxygen functionalization of the molecular precursor enables deterministic modulation of the electronic structure and magnetic ordering. This work establishes a general platform for designing correlated quantum materials with tunable flat band properties. △ Less

Submitted 18 June, 2026; originally announced June 2026.

Comments: arXiv admin note: substantial text overlap with arXiv:2411.01108

Back on Track: Aligning Rewards and States for Reasoning in Diffusion Large Language Models

Authors: Yawen Shao, Jie Xiao, Kai Zhu, Yu Liu, Hongchen Luo, Xueyang Fu, Yang Cao, Wei Zhai, Zheng-Jun Zha

Abstract: Reinforcement learning (RL) holds immense promise for enhancing the reasoning capabilities of diffusion large language models (dLLMs). However, progress is fundamentally constrained by a dual misalignment between authentic generation trajectory and the gradient update process: (i) Process-reward misalignment. Sparse, terminal rewards are indiscriminately assigned to all intermediate steps of the g… ▽ More Reinforcement learning (RL) holds immense promise for enhancing the reasoning capabilities of diffusion large language models (dLLMs). However, progress is fundamentally constrained by a dual misalignment between authentic generation trajectory and the gradient update process: (i) Process-reward misalignment. Sparse, terminal rewards are indiscriminately assigned to all intermediate steps of the generation process, failing to provide discriminative credit assignment. (ii) State-trajectory misalignment. Policy updates are often diverted toward artificial, out-of-trajectory states, squandering gradients on less informative samples. To address these limitations, we introduce Process Aligned Policy Optimization (PAPO), a novel framework that holistically aligns the RL update with the dLLM's generative trajectory via Step-Aware Process Rewards (SPR) that transform sparse terminal rewards into dense, step-wise credit, and Entropy-Guided Historical Re-enactment (EHR) that replays authentic trajectories at high-uncertainty steps. Extensive experiments on four benchmarks demonstrate that PAPO significantly outperforms baselines, achieving gains of up to 4.5% on GSM8K, 4.8% on MATH500, 42.2% on Countdown and 16.1% on Sudoku. △ Less

Submitted 7 June, 2026; originally announced June 2026.

Dual-Pathway Geometry-Aware MLLM for Spatial Intelligence

Authors: Yufei Zheng, Xuhan Zhu, Zide Liu, Chunpeng Zhou, Chenfeng Wang, Yongchao Xu, Yunnan Wang, Jiawei Liu, Pengfei Yu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Spatial understanding of the physical world from 2D visual inputs hinges on two complementary forms of geometric knowledge: holistic 3D structural perception and fine-grained metric scale estimation. Existing multimodal large language models (MLLMs) typically address only one facet, ingesting either depth maps or point clouds as additional model inputs, which incurs substantial computational overh… ▽ More Spatial understanding of the physical world from 2D visual inputs hinges on two complementary forms of geometric knowledge: holistic 3D structural perception and fine-grained metric scale estimation. Existing multimodal large language models (MLLMs) typically address only one facet, ingesting either depth maps or point clouds as additional model inputs, which incurs substantial computational overhead and inherits the generalization limitations of upstream prediction models. We propose GAMSI, a dual-pathway Geometry-Aware MLLM for Spatial Intelligence that takes only RGB images as input while internalizing both forms of geometric prior within a unified autoregressive backbone. Specifically, we introduce Metric-Structure Decoupled Queries (MSDQ) which employ two groups of learnable queries to respectively extract dense metric signals and sparse structural cues from the shared visual context, with a task-decoupled attention mask further preventing the two pathways from contaminating each other. Building on this, an Expert-Guided Visual Grounding (EVG) module projects the aggregated cues back to frame-level visual features and aligns them with vision foundation models, which serve purely as training-time supervision, rather than as model inputs. We further build a multi-task spatial instruction-tuning dataset (MTS) comprising 152{,}776 samples spanning 13 task types and three visual modalities, consolidated from six public datasets. Trained with a two-stage curriculum, GAMSI achieves state-of-the-art performance on seven spatial intelligence benchmarks. △ Less

Submitted 24 May, 2026; originally announced May 2026.

Self-Consistent Latent Reasoning: Long Latent Sequence Reasoning for Vision-Language Model

Authors: Chenfeng Wang, Wei He, Xuhan Zhu, Chunpeng Zhou, Qizhen Li, Song Yan, Yufei Zheng, Chengjun Yu, Fan Lu, Wei Zhai, Yang Cao, Pengfei Yu, Zheng-Jun Zha

Abstract: In language reasoning, longer chains of thought consistently yield better performance, which naturally suggests that visual latent reasoning may likewise benefit from longer latent sequences. However, we discover a counterintuitive phenomenon: the performance of existing latent visual reasoning methods systematically degrades as the latent sequence grows longer. We reveal the root cause: Informati… ▽ More In language reasoning, longer chains of thought consistently yield better performance, which naturally suggests that visual latent reasoning may likewise benefit from longer latent sequences. However, we discover a counterintuitive phenomenon: the performance of existing latent visual reasoning methods systematically degrades as the latent sequence grows longer. We reveal the root cause: Information Gain Collapse -- autoregressive generation makes each step highly dependent on prior outputs, so subsequent tokens can barely introduce new information. We further identify that heavily pooled ($\geq 128\times$) image embeddings used as supervision targets provide no more signal than meaningless placeholders. Motivated by these insights, we propose SCOLAR (Self-COnsistent LAtent Reasoning), which introduces a lightweight detransformer that leverages the LLM's full-sequence hidden states to generate auxiliary visual tokens in a single shot, with each token independently anchored to the original visual space. Combined with three-stage SFT and ALPO reinforcement learning, SCOLAR extends acceptable latent CoT length by over $30\times$, achieves state-of-the-art among open-source models on real-world reasoning benchmarks (+14.12% over backbone), and demonstrates strong out-of-distribution generalization. △ Less

Submitted 12 May, 2026; v1 submitted 12 May, 2026; originally announced May 2026.

Comments: 17 pages, 6 figures

Laplacian Frequency Interaction Network for Rural Thematic Road Extraction

Authors: Baiyan Chen, Weixin Zhai

Abstract: Rural thematic road network construction aims to extract topological road structures from movement trajectory images of agricultural machinery. However, this task faces challenges where downsampling methods commonly used in existing studies tend to blur the sparse high-frequency road structures, and the heavy noise from dense field operations often leads to fragmented or redundant topologies in th… ▽ More Rural thematic road network construction aims to extract topological road structures from movement trajectory images of agricultural machinery. However, this task faces challenges where downsampling methods commonly used in existing studies tend to blur the sparse high-frequency road structures, and the heavy noise from dense field operations often leads to fragmented or redundant topologies in the extracted networks. To address these challenges, we propose LFINet, a Laplacian Frequency Interaction Network. The network begins with a Laplacian Multi-scale Separator (LMS) to decouple the image into low-frequency semantic contexts and high-frequency structural details. These components are then processed by the Cross-Frequency Interaction Block (CFIB) through a dual-pathway architecture in which a High-Frequency Block (HFB) refines local structures while a Spatial Transformer (ST) captures global semantics. Subsequently, a Frequency Gated Modulation (FGM) mechanism integrates the features from pathways by leveraging semantic contexts to calibrate the structural details. Finally, a Progressive Reconstruction Decoder iteratively fuses multi-scale features to ensure topological consistency. Experiments conducted on a real-world agricultural trajectories dataset from Henan Province, China, show that LFINet establishes a new state-of-the-art. Specifically, it achieves an F1-score of 92.54% and an IoU of 86.12%, surpassing the second-ranked method by 0.64% and 1.1%, respectively. This confirms its capability to effectively construct topological road networks from noisy and sparse field data. △ Less

Submitted 4 May, 2026; originally announced May 2026.

AnchorSeg: Language Grounded Query Banks for Reasoning Segmentation

Authors: Rui Qian, Chuanhang Deng, Qiang Huang, Jian Xiong, Mingxuan Li, Yingbo Zhou, Wei Zhai, Jintao Chen, Dejing Dou

Abstract: Reasoning segmentation requires models to ground complex, implicit textual queries into precise pixel-level masks. Existing approaches rely on a single segmentation token $\texttt{<SEG>}$, whose hidden state implicitly encodes both semantic reasoning and spatial localization, limiting the model's ability to explicitly disentangle what to segment from where to segment. We introduce AnchorSeg, which… ▽ More Reasoning segmentation requires models to ground complex, implicit textual queries into precise pixel-level masks. Existing approaches rely on a single segmentation token $\texttt{<SEG>}$, whose hidden state implicitly encodes both semantic reasoning and spatial localization, limiting the model's ability to explicitly disentangle what to segment from where to segment. We introduce AnchorSeg, which reformulates reasoning segmentation as a structured conditional generation process over image tokens, conditioned on language grounded query banks. Instead of compressing all semantic reasoning and spatial localization into a single embedding, AnchorSeg constructs an ordered sequence of query banks: latent reasoning tokens that capture intermediate semantic states, and a segmentation anchor token that provides explicit spatial grounding. We model spatial conditioning as a factorized distribution over image tokens, where the anchor query determines localization signals while contextual queries provide semantic modulation. To bridge token-level predictions and pixel-level supervision, we propose Token--Mask Cycle Consistency (TMCC), a bidirectional training objective that enforces alignment across resolutions. By explicitly decoupling spatial grounding from semantic reasoning through structured language grounded query banks, AnchorSeg achieves state-of-the-art results on ReasonSeg test set (67.7\% gIoU and 68.1\% cIoU). All code and models are publicly available at https://github.com/rui-qian/AnchorSeg. △ Less

Submitted 21 April, 2026; v1 submitted 20 April, 2026; originally announced April 2026.

Comments: This work has been accepted to ACL 2026, please refer to https://github.com/rui-qian/AnchorSeg

MedPRMBench: A Fine-grained Benchmark for Process Reward Models in Medical Reasoning

Authors: Lingyan Wu, Xiang Zheng, Weiqi Zhai, Wei Wang, Xuan Ren, Zifan Zhang, Hu Wei, Bing Zhao

Abstract: Process-Level Reward Models (PRMs) are essential for guiding complex reasoning in large language models, yet existing PRM benchmarks cover only general domains such as mathematics, failing to address medical reasoning -- which is uniquely characterized by safety criticality, knowledge intensity, and diverse error patterns. Without a reliable medical PRM evaluation framework, we cannot quantify mod… ▽ More Process-Level Reward Models (PRMs) are essential for guiding complex reasoning in large language models, yet existing PRM benchmarks cover only general domains such as mathematics, failing to address medical reasoning -- which is uniquely characterized by safety criticality, knowledge intensity, and diverse error patterns. Without a reliable medical PRM evaluation framework, we cannot quantify models' error detection capabilities in clinical reasoning, leaving their safety in real-world healthcare applications unverified. We propose MedPRMBench, the first process-level reward model benchmark for the medical domain. Built through a three-phase pipeline based on Clinical Reasoning Blueprints (CRBs), MedPRMBench systematically generates high-quality evaluation data from seven medical QA sources, covering 14 fine-grained error types across three categories (Simplicity, Soundness, and Sensitivity) with the first 4-level severity grading system to quantify clinical impact. The benchmark comprises 6{,}500 questions with 13{,}000 reasoning chains and 113{,}910 step-level labels, plus 6{,}879 questions for training. Our medical PRM baseline achieves an 87.1\% overall PRMScore -- substantially surpassing all baselines -- and serves as a plug-and-play verifier that improves downstream medical QA accuracy by 3.2--6.7 percentage points. Systematic evaluation spanning proprietary frontier models, open-source reasoning models, and medical-specialized models reveals critical weaknesses in current models' medical reasoning error detection capabilities, providing clear directions for future PRM improvement. △ Less

Submitted 19 April, 2026; originally announced April 2026.

Gloria: Consistent Character Video Generation via Content Anchors

Authors: Yuhang Yang, Fan Zhang, Huaijin Pi, Shuai Guo, Guowei Xu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Digital characters are central to modern media, yet generating character videos with long-duration, consistent multi-view appearance and expressive identity remains challenging. Existing approaches either provide insufficient context to preserve identity or leverage non-character-centric information as the memory, leading to suboptimal consistency. Recognizing that character video generation inher… ▽ More Digital characters are central to modern media, yet generating character videos with long-duration, consistent multi-view appearance and expressive identity remains challenging. Existing approaches either provide insufficient context to preserve identity or leverage non-character-centric information as the memory, leading to suboptimal consistency. Recognizing that character video generation inherently resembles an outside-looking-in scenario. In this work, we propose representing the character visual attributes through a compact set of anchor frames. This design provides stable references for consistency, while reference-based video generation inherently faces challenges of copy-pasting and multi-reference conflicts. To address these, we introduce two mechanisms: Superset Content Anchoring, providing intra- and extra-training clip cues to prevent duplication, and RoPE as Weak Condition, encoding positional offsets to distinguish multiple anchors. Furthermore, we construct a scalable pipeline to extract these anchors from massive videos. Experiments show our method generates high-quality character videos exceeding 10 minutes, and achieves expressive identity and appearance consistency across views, surpassing existing methods. △ Less

Submitted 31 March, 2026; originally announced March 2026.

Comments: Accepted by CVPR2026 Main, project: https://yyvhang.github.io/Gloria_Page/

Automating Computational Chemistry Workflows via OpenClaw and Domain-Specific Skills

Authors: Mingwei Ding, Chen Huang, Yibo Hu, Yifan Li, Zitian Lu, Xingtai Yu, Duo Zhang, Wenxi Zhai, Tong Zhu, Qiangqiang Gu, Jinzhe Zeng

Abstract: This work presents a decoupled framework for multi-step computational chemistry automation built on OpenClaw. OpenClaw serves as the general-purpose agent for task coordination and supervision. Planning skills externalize task descriptions into executable task specifications, domain skills provide computational chemistry procedures, and the DPDispatcher skill grounds computation in heterogeneous H… ▽ More This work presents a decoupled framework for multi-step computational chemistry automation built on OpenClaw. OpenClaw serves as the general-purpose agent for task coordination and supervision. Planning skills externalize task descriptions into executable task specifications, domain skills provide computational chemistry procedures, and the DPDispatcher skill grounds computation in heterogeneous HPC environments. In a methane-oxidation reactive MD case study, the framework coordinated cross-tool execution, supported bounded recovery from runtime failures, and extracted reaction networks. △ Less

Submitted 12 May, 2026; v1 submitted 26 March, 2026; originally announced March 2026.

Comments: Rev 1

End-to-End Spatial-Temporal Transformer for Real-time 4D HOI Reconstruction

Authors: Haoyu Zhang, Wei Zhai, Yuhang Yang, Yang Cao, Zheng-Jun Zha

Abstract: Monocular 4D human-object interaction (HOI) reconstruction - recovering a moving human and a manipulated object from a single RGB video - remains challenging due to depth ambiguity and frequent occlusions. Existing methods often rely on multi-stage pipelines or iterative optimization, leading to high inference latency, failing to meet real-time requirements, and susceptibility to error accumulatio… ▽ More Monocular 4D human-object interaction (HOI) reconstruction - recovering a moving human and a manipulated object from a single RGB video - remains challenging due to depth ambiguity and frequent occlusions. Existing methods often rely on multi-stage pipelines or iterative optimization, leading to high inference latency, failing to meet real-time requirements, and susceptibility to error accumulation. To address these limitations, we propose THO, an end-to-end Spatial-Temporal Transformer that predicts human motion and coordinated object motion in a forward fashion from the given video and 3D template. THO achieves this by leveraging spatial-temporal HOI tuple priors. Spatial priors exploit contact-region proximity to infer occluded object features from human cues, while temporal priors capture cross-frame kinematic correlations to refine object representations and enforce physical coherence. Extensive experiments demonstrate that THO operates at an inference speed of 31.5 FPS on a single RTX 4090 GPU, achieving a >600x speedup over prior optimization-based methods while simultaneously improving reconstruction accuracy and temporal consistency. The project page is available at: https://nianheng.github.io/THO-project/ △ Less

Submitted 15 March, 2026; originally announced March 2026.

Comments: 23 pages, 7 figures. The project page is available at: https://nianheng.github.io/THO-project/

EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning

Authors: Chengjun Yu, Xuhan Zhu, Chaoqun Du, Pengfei Yu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Multimodal large language models (MLLMs) are increasingly considered as a foundation for embodied agents, yet it remains unclear whether they can reliably reason about the long-term physical consequences of actions from an egocentric viewpoint. We study this gap through a new task, Egocentric Scene Prediction with LOng-horizon REasoning: given an initial-scene image and a sequence of atomic action… ▽ More Multimodal large language models (MLLMs) are increasingly considered as a foundation for embodied agents, yet it remains unclear whether they can reliably reason about the long-term physical consequences of actions from an egocentric viewpoint. We study this gap through a new task, Egocentric Scene Prediction with LOng-horizon REasoning: given an initial-scene image and a sequence of atomic action descriptions, a model is asked to predict the final scene after all actions are executed. To enable systematic evaluation, we introduce EXPLORE-Bench, a benchmark curated from real first-person videos spanning diverse scenarios. Each instance pairs long action sequences with structured final-scene annotations, including object categories, visual attributes, and inter-object relations, which supports fine-grained, quantitative assessment. Experiments on a range of proprietary and open-source MLLMs reveal a significant performance gap to humans, indicating that long-horizon egocentric reasoning remains a major challenge. We further analyze test-time scaling via stepwise reasoning and show that decomposing long action sequences can improve performance to some extent, while incurring non-trivial computational overhead. Overall, EXPLORE-Bench provides a principled testbed for measuring and advancing long-horizon reasoning for egocentric embodied perception. △ Less

Submitted 26 June, 2026; v1 submitted 10 March, 2026; originally announced March 2026.

ClinConsensus: A Physician-Calibrated Benchmark for Evaluating Clinical Rubric Coverage in Chinese Medical LLMs

Authors: Xiang Zheng, Han Li, Wenjie Luo, Weiqi Zhai, Yiyuan Li, Chuanmiao Yan, Xue Yang, Kailuan Wu, Ruyi Xu, Tianyun Lu, Tianyi Tang, Yubo Ma, Kexin Yang, Dayiheng Liu, Sen Yang, Lin Qu, Bing Zhao, Hu Wei

Abstract: Open-ended medical LLM evaluation remains weakly grounded in physician-calibrated coverage of clinically relevant response criteria, especially in localized clinical settings. We introduce \textsc{ClinConsensus}, a Chinese medical benchmark of 2{,}500 expert-curated cases spanning 36 specialties, 12 task themes, multiple difficulty levels, and lay-facing versus professional-facing settings. Each c… ▽ More Open-ended medical LLM evaluation remains weakly grounded in physician-calibrated coverage of clinically relevant response criteria, especially in localized clinical settings. We introduce \textsc{ClinConsensus}, a Chinese medical benchmark of 2{,}500 expert-curated cases spanning 36 specialties, 12 task themes, multiple difficulty levels, and lay-facing versus professional-facing settings. Each case is paired with 30 case-specific binary rubric criteria. To evaluate whether responses satisfy enough physician-authored criteria, we propose \emph{Clinician-Anchored Coverage Score} (CACS), a physician-calibrated threshold metric instantiated at \(k=10\), and develop a dual-judge framework combining a GPT-5.1 grader with a physician-supervised Qwen3-8B judge. Evaluating 11 frontier LLMs, we find a persistent coverage gap: Rubric Accuracy ranges from 39.6\% to 52.1\%, whereas CACS@10 ranges from 17.8\% to 32.9\%, leaving a 19.2--21.9 point gap across models. Stratified analyses further reveal substantial variation across reasoning, evidence use, structured extraction, medication instructions, follow-up, and dialogue register. These results suggest that medical LLM evaluation should measure thresholded, rubric-grounded clinical coverage rather than average partial correctness. △ Less

Submitted 27 May, 2026; v1 submitted 2 March, 2026; originally announced March 2026.

EDU-MATRIX: A Society-Centric Generative Cognitive Digital Twin Architecture for Secondary Education

Authors: Wenjing Zhai, Jianbin Zhang, Tao Liu

Abstract: Existing multi-agent simulations often suffer from the "Agent-Centric Paradox": rules are hard-coded into individual agents, making complex social dynamics rigid and difficult to align with educational values. This paper presents EDU-MATRIX, a society-centric generative cognitive digital twin architecture that shifts the paradigm from simulating "people" to simulating a "social space with a gravit… ▽ More Existing multi-agent simulations often suffer from the "Agent-Centric Paradox": rules are hard-coded into individual agents, making complex social dynamics rigid and difficult to align with educational values. This paper presents EDU-MATRIX, a society-centric generative cognitive digital twin architecture that shifts the paradigm from simulating "people" to simulating a "social space with a gravitational field." We introduce three architectural contributions: (1) An Environment Context Injection Engine (ECIE), which acts as a "social microkernel," dynamically injecting institutional rules (Gravity) into agents based on their spatial-temporal coordinates; (2) A Modular Logic Evolution Protocol (MLEP), where knowledge exists as "fluid" capsules that agents synthesize to generate new paradigms, ensuring high dialogue consistency (94.1%); and (3) Endogenous Alignment via Role-Topology, where safety constraints emerge from the agent's position in the social graph rather than external filters. Deployed as a digital twin of a secondary school with 2,400 agents, the system demonstrates how "social gravity" (rules) and "cognitive fluids" (knowledge) interact to produce emergent, value-aligned behaviors (Social Clustering Coefficient: 0.72). △ Less

Submitted 20 February, 2026; originally announced February 2026.

Event-based Visual Deformation Measurement

Authors: Yuliang Wu, Wei Zhai, Yuxin Cui, Tiesong Zhao, Yang Cao, Zheng-Jun Zha

Abstract: Visual Deformation Measurement (VDM) aims to recover dense deformation fields by tracking surface motion from camera observations. Traditional image-based methods rely on minimal inter-frame motion to constrain the correspondence search space, which limits their applicability to highly dynamic scenes or necessitates high-speed cameras at the cost of prohibitive storage and computational overhead.… ▽ More Visual Deformation Measurement (VDM) aims to recover dense deformation fields by tracking surface motion from camera observations. Traditional image-based methods rely on minimal inter-frame motion to constrain the correspondence search space, which limits their applicability to highly dynamic scenes or necessitates high-speed cameras at the cost of prohibitive storage and computational overhead. We propose an event-frame fusion framework that exploits events for temporally dense motion cues and frames for spatially dense precise estimation. Revisiting the solid elastic modeling prior, we propose an Affine Invariant Simplicial (AIS) framework. It partitions the deformation field into linearized sub-regions with low-parametric representation, effectively mitigating motion ambiguities arising from sparse and noisy events. To speed up parameter searching and reduce error accumulation, a neighborhood-greedy optimization strategy is introduced, enabling well-converged sub-regions to guide their poorly-converged neighbors, effectively suppress local error accumulation in long-term dense tracking. To evaluate the proposed method, a benchmark dataset with temporally aligned event streams and frames is established, encompassing over 120 sequences spanning diverse deformation scenarios. Experimental results show that our method outperforms the state-of-the-art baseline by 1.6% in survival rate. Remarkably, it achieves this using only 18.9% of the data storage and processing resources of high-speed video methods. △ Less

Submitted 15 February, 2026; originally announced February 2026.

HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam

Authors: Weiqi Zhai, Zhihai Wang, Jinghang Wang, Boyu Yang, Xiaogang Li, Xander Xu, Bohan Wang, Peng Wang, Xingzhe Wu, Anfeng Li, Qiyuan Feng, Yuhao Zhou, Taolin Han, Wenjie Luo, Yiyuan Li, Xiang Zheng, Yaxuan Wang, Ruixiang Luo, Guojie Lin, Peiyao Xiao, Chengliang Xu, Ben Wang, Zeyu Wang, Zichao Chen, Jianan Ye , et al. (11 additional authors not shown)

Abstract: Humanity's Last Exam (HLE) has become a widely used benchmark for evaluating frontier large language models on challenging, multi-domain questions. However, community-led analyses have raised concerns that HLE contains a non-trivial number of noisy items, which can bias evaluation results and distort cross-model comparisons. To address this challenge, we introduce HLE-Verified, a verified and revi… ▽ More Humanity's Last Exam (HLE) has become a widely used benchmark for evaluating frontier large language models on challenging, multi-domain questions. However, community-led analyses have raised concerns that HLE contains a non-trivial number of noisy items, which can bias evaluation results and distort cross-model comparisons. To address this challenge, we introduce HLE-Verified, a verified and revised version of HLE with a transparent verification protocol and fine-grained error taxonomy. Our construction follows a two-stage validation-and-repair workflow resulting in a certified benchmark. In Stage I, each item undergoes binary validation of the problem and final answer through domain-expert review and model-based cross-checks, yielding 668 verified items. In Stage II, flawed but fixable items are revised under strict constraints preserving the original evaluation intent, through dual independent expert repairs, model-assisted auditing, and final adjudication, resulting in 1,143 revised-and-certified items. The remaining 689 items are released as a documented uncertain set with explicit uncertainty sources and expertise tags for future refinement. We evaluate eight state-of-the-art language models on HLE and HLE-Verified, observing an average absolute accuracy gain of 7--10 percentage points on HLE-Verified. The improvement is particularly pronounced on items where the original problem statement and/or reference answer is erroneous, with gains of 30--40 percentage points. Our analyses further reveal a strong association between model confidence and the presence of errors in the problem statement or reference answer, supporting the effectiveness of our revisions. Overall, HLE-Verified improves HLE-style evaluations by reducing annotation noise and enabling more faithful measurement of model capabilities. Data is available at: https://huggingface.co/datasets/skylenage/HLE-Verified △ Less

Submitted 17 August, 2026; v1 submitted 14 February, 2026; originally announced February 2026.

Comments: 14 pages, 10 figures

Unbiased Gradient Estimation for Event Binning via Functional Backpropagation

Authors: Jinze Chen, Wei Zhai, Han Han, Tiankai Ma, Yang Cao, Bin Li, Zheng-Jun Zha

Abstract: Event-based vision encodes dynamic scenes as asynchronous spatio-temporal spikes called events. To leverage conventional image processing pipelines, events are typically binned into frames. However, binning functions are discontinuous, which truncates gradients at the frame level and forces most event-based algorithms to rely solely on frame-based features. Attempts to directly learn from raw even… ▽ More Event-based vision encodes dynamic scenes as asynchronous spatio-temporal spikes called events. To leverage conventional image processing pipelines, events are typically binned into frames. However, binning functions are discontinuous, which truncates gradients at the frame level and forces most event-based algorithms to rely solely on frame-based features. Attempts to directly learn from raw events avoid this restriction but instead suffer from biased gradient estimation due to the discontinuities of the binning operation, ultimately limiting their learning efficiency. To address this challenge, we propose a novel framework for unbiased gradient estimation of arbitrary binning functions by synthesizing weak derivatives during backpropagation while keeping the forward output unchanged. The key idea is to exploit integration by parts: lifting the target functions to functionals yields an integral form of the derivative of the binning function during backpropagation, where the cotangent function naturally arises. By reconstructing this cotangent function from the sampled cotangent vector, we compute weak derivatives that provably match long-range finite differences of both smooth and non-smooth targets. Experimentally, our method improves simple optimization-based egomotion estimation with 3.2\% lower RMS error and 1.57$\times$ faster convergence. On complex downstream tasks, we achieve 9.4\% lower EPE in self-supervised optical flow, and 5.1\% lower RMS error in SLAM, demonstrating broad benefits for event-based visual perception. Source code can be found at https://github.com/chjz1024/EventFBP. △ Less

Submitted 12 February, 2026; originally announced February 2026.

Unmasking Reasoning Processes: A Process-aware Benchmark for Evaluating Structural Mathematical Reasoning in LLMs

Authors: Xiang Zheng, Weiqi Zhai, Wei Wang, Boyu Yang, Wenbo Li, Ruixiang Luo, Haoxiang Sun, Yucheng Wang, Zhengze Li, Meng Wang, Yuetian Du, Guojie Lin, Yaxuan Wang, Xiaoxiao Xu, Yanhu Mo, Xuan Ren, Hu Wei, Bing Zhao

Abstract: Recent large language models (LLMs) achieve near-saturation accuracy on many established mathematical reasoning benchmarks, raising concerns about their ability to diagnose genuine reasoning competence. This saturation largely stems from the dominance of template-based computation and shallow arithmetic decomposition in existing datasets, which underrepresent reasoning skills such as multi-constra… ▽ More Recent large language models (LLMs) achieve near-saturation accuracy on many established mathematical reasoning benchmarks, raising concerns about their ability to diagnose genuine reasoning competence. This saturation largely stems from the dominance of template-based computation and shallow arithmetic decomposition in existing datasets, which underrepresent reasoning skills such as multi-constraint coordination, constructive logical synthesis, and spatial inference. To address this gap, we introduce ReasoningMath-Plus, a benchmark of 150 carefully curated problems explicitly designed to evaluate structural reasoning. Each problem emphasizes reasoning under interacting constraints, constructive solution formation, or non-trivial structural insight, and is annotated with a minimal reasoning skeleton to support fine-grained process-level evaluation. Alongside the dataset, we introduce HCRS (Hazard-aware Chain-based Rule Score), a deterministic step-level scoring function, and train a Process Reward Model (PRM) on the annotated reasoning traces. Empirically, while leading models attain relatively high final-answer accuracy (up to 5.8/10), HCRS-based holistic evaluation yields substantially lower scores (average 4.36/10, best 5.14/10), showing that answer-only metrics can overestimate reasoning robustness. △ Less

Submitted 26 February, 2026; v1 submitted 31 January, 2026; originally announced February 2026.

Comments: 8 pages, and 3 figures

TrackTeller: Temporal Multimodal 3D Grounding for Behavior-Dependent Object References

Authors: Jiahong Yu, Ziqi Wang, Hailiang Zhao, Wei Zhai, Xueqiang Yan, Shuiguang Deng

Abstract: Understanding natural-language references to objects in dynamic 3D driving scenes is essential for interactive autonomous systems. In practice, many referring expressions describe targets through recent motion or short-term interactions, which cannot be resolved from static appearance or geometry alone. We study temporal language-based 3D grounding, where the objective is to identify the referred… ▽ More Understanding natural-language references to objects in dynamic 3D driving scenes is essential for interactive autonomous systems. In practice, many referring expressions describe targets through recent motion or short-term interactions, which cannot be resolved from static appearance or geometry alone. We study temporal language-based 3D grounding, where the objective is to identify the referred object in the current frame by leveraging multi-frame observations. We propose TrackTeller, a temporal multimodal grounding framework that integrates LiDAR-image fusion, language-conditioned decoding, and temporal reasoning in a unified architecture. TrackTeller constructs a shared UniScene representation aligned with textual semantics, generates language-aware 3D proposals, and refines grounding decisions using motion history and short-term dynamics. Experiments on the NuPrompt benchmark demonstrate that TrackTeller consistently improves language-grounded tracking performance, outperforming strong baselines with a 70% relative improvement in Average Multi-Object Tracking Accuracy and a 3.15-3.4 times reduction in False Alarm Frequency. △ Less

Submitted 25 December, 2025; originally announced December 2025.

MatE: Material Extraction from Single-Image via Geometric Prior

Authors: Zeyu Zhang, Wei Zhai, Jian Yang, Yang Cao

Abstract: The creation of high-fidelity, physically-based rendering (PBR) materials remains a bottleneck in many graphics pipelines, typically requiring specialized equipment and expert-driven post-processing. To democratize this process, we present MatE, a novel method for generating tileable PBR materials from a single image taken under unconstrained, real-world conditions. Given an image and a user-provi… ▽ More The creation of high-fidelity, physically-based rendering (PBR) materials remains a bottleneck in many graphics pipelines, typically requiring specialized equipment and expert-driven post-processing. To democratize this process, we present MatE, a novel method for generating tileable PBR materials from a single image taken under unconstrained, real-world conditions. Given an image and a user-provided mask, MatE first performs coarse rectification using an estimated depth map as a geometric prior, and then employs a dual-branch diffusion model. Leveraging a learned consistency from rotation-aligned and scale-aligned training data, this model further rectify residual distortions from the coarse result and translate it into a complete set of material maps, including albedo, normal, roughness and height. Our framework achieves invariance to the unknown illumination and perspective of the input image, allowing for the recovery of intrinsic material properties from casual captures. Through comprehensive experiments on both synthetic and real-world data, we demonstrate the efficacy and robustness of our approach, enabling users to create realistic materials from real-world image. △ Less

Submitted 20 December, 2025; originally announced December 2025.

Comments: 8 pages, 8 figures

Anchoring Values in Temporal and Group Dimensions for Flow Matching Model Alignment

Authors: Yawen Shao, Jie Xiao, Kai Zhu, Yu Liu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Group Relative Policy Optimization (GRPO) has proven highly effective in enhancing the alignment capabilities of Large Language Models (LLMs). However, current adaptations of GRPO for the flow matching-based image generation neglect a foundational conflict between its core principles and the distinct dynamics of the visual synthesis process. This mismatch leads to two key limitations: (i) Uniforml… ▽ More Group Relative Policy Optimization (GRPO) has proven highly effective in enhancing the alignment capabilities of Large Language Models (LLMs). However, current adaptations of GRPO for the flow matching-based image generation neglect a foundational conflict between its core principles and the distinct dynamics of the visual synthesis process. This mismatch leads to two key limitations: (i) Uniformly applying a sparse terminal reward across all timesteps impairs temporal credit assignment, ignoring the differing criticality of generation phases from early structure formation to late-stage tuning. (ii) Exclusive reliance on relative, intra-group rewards causes the optimization signal to fade as training converges, leading to the optimization stagnation when reward diversity is entirely depleted. To address these limitations, we propose Value-Anchored Group Policy Optimization (VGPO), a framework that redefines value estimation across both temporal and group dimensions. Specifically, VGPO transforms the sparse terminal reward into dense, process-aware value estimates, enabling precise credit assignment by modeling the expected cumulative reward at each generative stage. Furthermore, VGPO replaces standard group normalization with a novel process enhanced by absolute values to maintain a stable optimization signal even as reward diversity declines. Extensive experiments on three benchmarks demonstrate that VGPO achieves state-of-the-art image quality while simultaneously improving task-specific accuracy, effectively mitigating reward hacking. Project webpage: https://yawen-shao.github.io/VGPO/. △ Less

Submitted 13 December, 2025; originally announced December 2025.

WeMMU: Enhanced Bridging of Vision-Language Models and Diffusion Models via Noisy Query Tokens

Authors: Jian Yang, Dacheng Yin, Xiaoxuan He, Yong Li, Fengyun Rao, Jing Lyu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Recent progress in multimodal large language models (MLLMs) has highlighted the challenge of efficiently bridging pre-trained Vision-Language Models (VLMs) with Diffusion Models. While methods using a fixed number of learnable query tokens offer computational efficiency, they suffer from task generalization collapse, failing to adapt to new tasks that are distant from their pre-training tasks. To… ▽ More Recent progress in multimodal large language models (MLLMs) has highlighted the challenge of efficiently bridging pre-trained Vision-Language Models (VLMs) with Diffusion Models. While methods using a fixed number of learnable query tokens offer computational efficiency, they suffer from task generalization collapse, failing to adapt to new tasks that are distant from their pre-training tasks. To overcome this, we propose Noisy Query Tokens, which learn a distributed representation space between the VLM and Diffusion Model via end-to-end optimization, enhancing continual learning. Additionally, we introduce a VAE branch with linear projection to recover fine-grained image details. Experimental results confirm our approach mitigates generalization collapse and enables stable continual learning across diverse tasks. △ Less

Submitted 2 December, 2025; originally announced December 2025.

Determinism of Randomness: Prompt-Residual Seed Shaping for Diffusion Generation

Authors: Song Yan, Wei Zhai, Chenfeng Wang, Xinliang Bi, Jian Yang, Yancheng Cai, Yusen Zhang, Yunwei Lan, Tao Zhang, GuanYe Xiong, Min Li, Zheng-Jun Zha

Abstract: Diffusion models start generation from an isotropic Gaussian latent, yet changing only the random seed can lead to large differences in prompt faithfulness, composition, and visual quality. We study this seed sensitivity through the semantic map from initial noise to generated meaning. Although the sampling flow is locally invertible, the subsequent semantic projection is many-to-one, inducing a d… ▽ More Diffusion models start generation from an isotropic Gaussian latent, yet changing only the random seed can lead to large differences in prompt faithfulness, composition, and visual quality. We study this seed sensitivity through the semantic map from initial noise to generated meaning. Although the sampling flow is locally invertible, the subsequent semantic projection is many-to-one, inducing a degenerate pullback semi-metric on the latent space: most local directions are nearly semantic-invariant, while semantic-sensitive variation is concentrated in a much smaller horizontal subspace. This provides an explanatory geometric view of the seed lottery. Motivated by this view, we introduce a training-free prompt-residual seed-shaping procedure. Rather than claiming to recover the exact horizontal space, the method uses a single high-noise cold-start prompt residual as a model-coupled proxy, injects only its tangential component, and retracts the seed to the original Gaussian radius shell. This keeps the initialization prior-compatible while adding only one conditional/unconditional probe before standard sampling. Across multiple generation benchmarks, the method improves alignment and quality metrics over standard sampling, supporting both the practical value of the proxy and the explanatory relevance of semantic anisotropy. △ Less

Submitted 11 May, 2026; v1 submitted 10 November, 2025; originally announced November 2025.

Oral Tradition-Encoded NanyinHGNN: Integrating Nanyin Music Preservation and Generation through a Pipa-Centric Dataset

Authors: Jianbing Xiahou, Weixi Zhai, Xu Cui

Abstract: We propose NanyinHGNN, a heterogeneous graph network model for generating Nanyin instrumental music. As a UNESCO-recognized intangible cultural heritage, Nanyin follows a heterophonic tradition centered around the pipa, where core melodies are notated in traditional notation while ornamentations are passed down orally, presenting challenges for both preservation and contemporary innovation. To add… ▽ More We propose NanyinHGNN, a heterogeneous graph network model for generating Nanyin instrumental music. As a UNESCO-recognized intangible cultural heritage, Nanyin follows a heterophonic tradition centered around the pipa, where core melodies are notated in traditional notation while ornamentations are passed down orally, presenting challenges for both preservation and contemporary innovation. To address this, we construct a Pipa-Centric MIDI dataset, develop NanyinTok as a specialized tokenization method, and convert symbolic sequences into graph structures using a Graph Converter to ensure that key musical features are preserved. Our key innovation reformulates ornamentation generation as the creation of ornamentation nodes within a heterogeneous graph. First, a graph neural network generates melodic outlines optimized for ornamentations. Then, a rule-guided system informed by Nanyin performance practices refines these outlines into complete ornamentations without requiring explicit ornamentation annotations during training. Experimental results demonstrate that our model successfully generates authentic heterophonic ensembles featuring four traditional instruments. These findings validate that integrating domain-specific knowledge into model architecture can effectively mitigate data scarcity challenges in computational ethnomusicology. △ Less

Submitted 28 October, 2025; originally announced October 2025.

Comments: 10 pages, 2 figures

TOUCH: Text-guided Controllable Generation of Free-Form Hand-Object Interactions

Authors: Guangyi Han, Wei Zhai, Yuhang Yang, Yang Cao, Zheng-Jun Zha

Abstract: Hand-object interaction (HOI) is fundamental for humans to express intent. Existing HOI generation research is predominantly confined to fixed grasping patterns, where control is tied to physical priors such as force closure or generic intent instructions, even when expressed through elaborate language. Such an overly general conditioning imposes a strong inductive bias for stable grasps, thus fai… ▽ More Hand-object interaction (HOI) is fundamental for humans to express intent. Existing HOI generation research is predominantly confined to fixed grasping patterns, where control is tied to physical priors such as force closure or generic intent instructions, even when expressed through elaborate language. Such an overly general conditioning imposes a strong inductive bias for stable grasps, thus failing to capture the diversity of daily HOI. To address these limitations, we introduce Free-Form HOI Generation, which aims to generate controllable, diverse, and physically plausible HOI conditioned on fine-grained intent, extending HOI from grasping to free-form interactions, like pushing, poking, and rotating. To support this task, we construct WildO2, an in-the-wild diverse 3D HOI dataset, which includes diverse HOI derived from internet videos. Specifically, it contains 4.4k unique interactions across 92 intents and 610 object categories, each with detailed semantic annotations. Building on this dataset, we propose TOUCH, a three-stage framework centered on a multi-level diffusion model that facilitates fine-grained semantic control to generate versatile hand poses beyond grasping priors. This process leverages explicit contact modeling for conditioning and is subsequently refined with contact consistency and physical constraints to ensure realism. Comprehensive experiments demonstrate our method's ability to generate controllable, diverse, and physically plausible hand interactions representative of daily activities. The project page is $\href{https://guangyid.github.io/hoi123touch}{here}$. △ Less

Submitted 16 October, 2025; originally announced October 2025.

UGround: Towards Unified Visual Grounding with Unrolled Transformers

Authors: Rui Qian, Xin Yin, Chuanhang Deng, Zhiyuan Peng, Jian Xiong, Wei Zhai, Dejing Dou

Abstract: We present UGround, a \textbf{U}nified visual \textbf{Ground}ing paradigm that dynamically selects intermediate layers across \textbf{U}nrolled transformers as ``mask as prompt,'' diverging from the prevailing pipeline that leverages the fixed last hidden layer as ``\texttt{<SEG>} as prompt.'' UGround addresses two primary challenges posed by the prevailing paradigm: (1) its reliance on the fixed… ▽ More We present UGround, a \textbf{U}nified visual \textbf{Ground}ing paradigm that dynamically selects intermediate layers across \textbf{U}nrolled transformers as ``mask as prompt,'' diverging from the prevailing pipeline that leverages the fixed last hidden layer as ``\texttt{<SEG>} as prompt.'' UGround addresses two primary challenges posed by the prevailing paradigm: (1) its reliance on the fixed last hidden layer, which sequentially amplifies cumulative errors arising from layer-by-layer propagation without intermediate correction, and (2) its use of \texttt{<SEG>} as a prompt, which implicitly projects textual embeddings into visual space without explicit spatial cues (e.g., coordinates). Central to UGround is Policy-Prompted Masking, which comprises two key components: Stochastic Skip Connection (SSC) and Mask as Prompt (MasP). SSC is a reinforcement learning policy that, via stochastic sampling, allows each \texttt{<SEG>} token to slide across unrolled transformer layers, enabling dynamic layer selection at which it connects to the vision model (e.g., SAM) in a skip-connection fashion. Given the selected hidden layer, MasP uses the similarity map derived from the \texttt{<SEG>} token and image tokens as a soft logit mask to prompt SAM for mask generation, offering explicit spatial cues through its activation regions. To validate the effectiveness of UGround, we, for the first time, have unified visual grounding within a single framework from an attribute perspective, spanning from traditional refer expression segmentation to newly proposed reasoning segmentation, single-target to multi-target, positive query to false premise (empty target). All code and models are publicly available at https://github.com/rui-qian/UGround. △ Less

Submitted 12 May, 2026; v1 submitted 4 October, 2025; originally announced October 2025.

Comments: This work has been accepted to ICML 2026, please refer to https://github.com/rui-qian/UGround

SKYLENAGE Technical Report: Mathematical Reasoning and Contest-Innovation Benchmarks for Multi-Level Math Evaluation

Authors: Hu Wei, Ze Xu, Boyu Yang, Linlin Miao, Weiqi Zhai, Yihan Li, Zixuan Li, Zhijun Wang, Boya Wang, Jianwei Yu, Jialing Yuan, Xiaoyue Zhang, Cheng He, Minglei Chen, Zifan Zhang, Qianhui Li, Wei Wang, Xiang Xu

Abstract: Large language models (LLMs) now perform strongly on many public math suites, yet frontier separation within mathematics increasingly suffers from ceiling effects. We present two complementary benchmarks: SKYLENAGE-ReasoningMATH, a 100-item, structure-aware diagnostic set with per-item metadata on length, numeric density, and symbolic complexity; and SKYLENAGE-MATH, a 150-item contest-style suite… ▽ More Large language models (LLMs) now perform strongly on many public math suites, yet frontier separation within mathematics increasingly suffers from ceiling effects. We present two complementary benchmarks: SKYLENAGE-ReasoningMATH, a 100-item, structure-aware diagnostic set with per-item metadata on length, numeric density, and symbolic complexity; and SKYLENAGE-MATH, a 150-item contest-style suite spanning four stages from high school to doctoral under a seven-subject taxonomy. We evaluate fifteen contemporary LLM variants under a single setup and analyze subject x model and grade x model performance. On the contest suite, the strongest model reaches 44% while the runner-up reaches 37%; accuracy declines from high school to doctoral, and top systems exhibit a doctoral-to-high-school retention near 79%. On the reasoning set, the best model attains 81% overall, and hardest-slice results reveal clear robustness gaps between leaders and the mid-tier. In summary, we release SKYLENAGE-ReasoningMATH and report aggregate results for SKYLENAGE-MATH; together, SKYLENAGE provides a hard, reasoning-centered and broadly covering math benchmark with calibrated difficulty and rich metadata, serving as a reference benchmark for future evaluations of mathematical reasoning. △ Less

Submitted 23 September, 2025; originally announced October 2025.

A Spatio-Temporal Feature Fusion EEG Virtual Channel Signal Generation Network and Its Application in Anxiety Assessment

Authors: Shangqing Yuan, Wenshuang Zhai, Shengwen Guo

Abstract: To address the issue of limited channels and insufficient information collection in portable EEG devices, this study explores an EEG virtual channel signal generation network using a novel spatio-temporal feature fusion strategy. Based on the EEG signals from four frontal lobe channels, the network aims to generate virtual channel EEG signals for other 13 important brain regions. The architecture… ▽ More To address the issue of limited channels and insufficient information collection in portable EEG devices, this study explores an EEG virtual channel signal generation network using a novel spatio-temporal feature fusion strategy. Based on the EEG signals from four frontal lobe channels, the network aims to generate virtual channel EEG signals for other 13 important brain regions. The architecture of the network is a two-dimensional convolutional neural network and it includes a parallel module for temporal and spatial domain feature extraction, followed by a feature fusion module. The public PRED+CT database, which includes multi-channel EEG signals from 119 subjects, was selected to verify the constructed network. The results showed that the average correlation coefficient between the generated virtual channel EEG signals and the original real signals was 0.6724, with an average absolute error of 3.9470. Furthermore, the 13 virtual channel EEG signals were combined with the original EEG signals of four brain regions and then used for anxiety classification with a support vector machine. The results indicate that the virtual EEG signals generated by the constructed network not only have a high degree of consistency with the real channel EEG signals but also significantly enhance the performance of machine learning algorithms for anxiety classification. This study effectively alleviates the problem of insufficient information acquisition by portable EEG devices with few channels. △ Less

Submitted 14 September, 2025; originally announced September 2025.

GRPOformer: Advancing Hyperparameter Optimization via Group Relative Policy Optimization

Authors: Haoxin Guo, Jiawen Pan, Weixin Zhai

Abstract: Hyperparameter optimization (HPO) plays a critical role in improving model performance. Transformer-based HPO methods have shown great potential; however, existing approaches rely heavily on large-scale historical optimization trajectories and lack effective reinforcement learning (RL) techniques, thereby limiting their efficiency and performance improvements. Inspired by the success of Group Rela… ▽ More Hyperparameter optimization (HPO) plays a critical role in improving model performance. Transformer-based HPO methods have shown great potential; however, existing approaches rely heavily on large-scale historical optimization trajectories and lack effective reinforcement learning (RL) techniques, thereby limiting their efficiency and performance improvements. Inspired by the success of Group Relative Policy Optimization (GRPO) in large language models (LLMs), we propose GRPOformer -- a novel hyperparameter optimization framework that integrates reinforcement learning (RL) with Transformers. In GRPOformer, Transformers are employed to generate new hyperparameter configurations from historical optimization trajectories, while GRPO enables rapid trajectory construction and optimization strategy learning from scratch. Moreover, we introduce Policy Churn Regularization (PCR) to enhance the stability of GRPO training. Experimental results on OpenML demonstrate that GRPOformer consistently outperforms baseline methods across diverse tasks, offering new insights into the application of RL for HPO. △ Less

Submitted 21 September, 2025; originally announced September 2025.

Floating-Body Hydrodynamic Neural Networks

Authors: Tianshuo Zhang, Wenzhe Zhai, Rui Yann, Jia Gao, He Cao, Xianglei Xing

Abstract: Fluid-structure interaction is common in engineering and natural systems, where floating-body motion is governed by added mass, drag, and background flows. Modeling these dissipative dynamics is difficult: black-box neural models regress state derivatives with limited interpretability and unstable long-horizon predictions. We propose Floating-Body Hydrodynamic Neural Networks (FHNN), a physics-str… ▽ More Fluid-structure interaction is common in engineering and natural systems, where floating-body motion is governed by added mass, drag, and background flows. Modeling these dissipative dynamics is difficult: black-box neural models regress state derivatives with limited interpretability and unstable long-horizon predictions. We propose Floating-Body Hydrodynamic Neural Networks (FHNN), a physics-structured framework that predicts interpretable hydrodynamic parameters such as directional added masses, drag coefficients, and a streamfunction-based flow, and couples them with analytic equations of motion. This design constrains the hypothesis space, enhances interpretability, and stabilizes integration. On synthetic vortex datasets, FHNN achieves up to an order-of-magnitude lower error than Neural ODEs, recovers physically consistent flow fields. Compared with Hamiltonian and Lagrangian neural networks, FHNN more effectively handles dissipative dynamics while preserving interpretability, which bridges the gap between black-box learning and transparent system identification. △ Less

Submitted 17 September, 2025; originally announced September 2025.

ViewPoint: Panoramic Video Generation with Pretrained Diffusion Models

Authors: Zixun Fang, Kai Zhu, Zhiheng Liu, Yu Liu, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Panoramic video generation aims to synthesize 360-degree immersive videos, holding significant importance in the fields of VR, world models, and spatial intelligence. Existing works fail to synthesize high-quality panoramic videos due to the inherent modality gap between panoramic data and perspective data, which constitutes the majority of the training data for modern diffusion models. In this pa… ▽ More Panoramic video generation aims to synthesize 360-degree immersive videos, holding significant importance in the fields of VR, world models, and spatial intelligence. Existing works fail to synthesize high-quality panoramic videos due to the inherent modality gap between panoramic data and perspective data, which constitutes the majority of the training data for modern diffusion models. In this paper, we propose a novel framework utilizing pretrained perspective video models for generating panoramic videos. Specifically, we design a novel panorama representation named ViewPoint map, which possesses global spatial continuity and fine-grained visual details simultaneously. With our proposed Pano-Perspective attention mechanism, the model benefits from pretrained perspective priors and captures the panoramic spatial correlations of the ViewPoint map effectively. Extensive experiments demonstrate that our method can synthesize highly dynamic and spatially consistent panoramic videos, achieving state-of-the-art performance and surpassing previous methods. △ Less

Submitted 30 June, 2025; originally announced June 2025.

Comments: https://becauseimbatman0.github.io/ViewPoint

MUCAR: Benchmarking Multilingual Cross-Modal Ambiguity Resolution for Multimodal Large Language Models

Authors: Xiaolong Wang, Zhaolu Kang, Wangyuxuan Zhai, Xinyue Lou, Yunghwei Lai, Ziyue Wang, Yawen Wang, Kaiyu Huang, Yile Wang, Peng Li, Yang Liu

Abstract: Multimodal Large Language Models (MLLMs) have demonstrated significant advances across numerous vision-language tasks. MLLMs have shown promising capability in aligning visual and textual modalities, allowing them to process image-text pairs with clear and explicit meanings. However, resolving the inherent ambiguities present in real-world language and visual contexts remains a challenge. Existing… ▽ More Multimodal Large Language Models (MLLMs) have demonstrated significant advances across numerous vision-language tasks. MLLMs have shown promising capability in aligning visual and textual modalities, allowing them to process image-text pairs with clear and explicit meanings. However, resolving the inherent ambiguities present in real-world language and visual contexts remains a challenge. Existing multimodal benchmarks typically overlook linguistic and visual ambiguities, relying mainly on unimodal context for disambiguation and thus failing to exploit the mutual clarification potential between modalities. To bridge this gap, we introduce MUCAR, a novel and challenging benchmark designed explicitly for evaluating multimodal ambiguity resolution across multilingual and cross-modal scenarios. MUCAR includes first a multilingual dataset where ambiguous textual expressions are uniquely resolved by corresponding visual contexts, and second a dual-ambiguity dataset that systematically pairs ambiguous images with ambiguous textual contexts, with each combination carefully constructed to yield a single, clear interpretation through mutual disambiguation. Extensive evaluations involving 19 state-of-the-art multimodal models--encompassing both open-source and proprietary architectures--reveal substantial gaps compared to human-level performance, highlighting the need for future research into more sophisticated cross-modal ambiguity comprehension methods, further pushing the boundaries of multimodal reasoning. △ Less

Submitted 26 September, 2025; v1 submitted 20 June, 2025; originally announced June 2025.

Heterogeneous Secure Transmissions in IRS-Assisted NOMA Communications: CO-GNN Approach

Authors: Linlin Liang, Zongkai Tian, Haiyan Huang, Xiaoyan Li, Zhisheng Yin, Dehua Zhang, Nina Zhang, Wenchao Zhai

Abstract: Intelligent Reflecting Surfaces (IRS) enhance spectral efficiency by adjusting reflection phase shifts, while Non-Orthogonal Multiple Access (NOMA) increases system capacity. Consequently, IRS-assisted NOMA communications have garnered significant research interest. However, the passive nature of the IRS, lacking authentication and security protocols, makes these systems vulnerable to external eav… ▽ More Intelligent Reflecting Surfaces (IRS) enhance spectral efficiency by adjusting reflection phase shifts, while Non-Orthogonal Multiple Access (NOMA) increases system capacity. Consequently, IRS-assisted NOMA communications have garnered significant research interest. However, the passive nature of the IRS, lacking authentication and security protocols, makes these systems vulnerable to external eavesdropping due to the openness of electromagnetic signal propagation and reflection. NOMA's inherent multi-user signal superposition also introduces internal eavesdropping risks during user pairing. This paper investigates secure transmissions in IRS-assisted NOMA systems with heterogeneous resource configuration in wireless networks to mitigate both external and internal eavesdropping. To maximize the sum secrecy rate of legitimate users, we propose a combinatorial optimization graph neural network (CO-GNN) approach to jointly optimize beamforming at the base station, power allocation of NOMA users, and phase shifts of IRS for dynamic heterogeneous resource allocation, thereby enabling the design of dual-link or multi-link secure transmissions in the presence of eavesdroppers on the same or heterogeneous links. The CO-GNN algorithm simplifies the complex mathematical problem-solving process, eliminates the need for channel estimation, and enhances scalability. Simulation results demonstrate that the proposed algorithm significantly enhances the secure transmission performance of the system. △ Less

Submitted 3 June, 2025; originally announced June 2025.

Towards Sequence Modeling Alignment between Tokenizer and Autoregressive Model

Authors: Pingyu Wu, Kai Zhu, Yu Liu, Longxiang Tang, Jian Yang, Yansong Peng, Wei Zhai, Yang Cao, Zheng-Jun Zha

Abstract: Autoregressive image generation aims to predict the next token based on previous ones. However, this process is challenged by the bidirectional dependencies inherent in conventional image tokenizations, which creates a fundamental misalignment with the unidirectional nature of autoregressive models. To resolve this, we introduce AliTok, a novel Aligned Tokenizer that alters the dependency structur… ▽ More Autoregressive image generation aims to predict the next token based on previous ones. However, this process is challenged by the bidirectional dependencies inherent in conventional image tokenizations, which creates a fundamental misalignment with the unidirectional nature of autoregressive models. To resolve this, we introduce AliTok, a novel Aligned Tokenizer that alters the dependency structure of the token sequence. AliTok employs a bidirectional encoder constrained by a causal decoder, a design that compels the encoder to produce a token sequence with both semantic richness and forward-dependency. Furthermore, by incorporating prefix tokens and employing a two-stage tokenizer training process to enhance reconstruction performance, AliTok achieves high fidelity and predictability simultaneously. Building upon AliTok, a standard decoder-only autoregressive model with just 177M parameters achieves a gFID of 1.44 and an IS of 319.5 on ImageNet-256. Scaling to 662M, our model reaches a gFID of 1.28, surpassing the SOTA diffusion method with 10x faster sampling. On ImageNet-512, our 318M model also achieves a SOTA gFID of 1.39. Code and weights at https://github.com/ali-vilab/alitok. △ Less

Submitted 15 February, 2026; v1 submitted 5 June, 2025; originally announced June 2025.

Comments: ICLR2026

Towards Better De-raining Generalization via Rainy Characteristics Memorization and Replay

Authors: Kunyu Wang, Xueyang Fu, Chengzhi Cao, Chengjie Ge, Wei Zhai, Zheng-Jun Zha

Abstract: Current image de-raining methods primarily learn from a limited dataset, leading to inadequate performance in varied real-world rainy conditions. To tackle this, we introduce a new framework that enables networks to progressively expand their de-raining knowledge base by tapping into a growing pool of datasets, significantly boosting their adaptability. Drawing inspiration from the human brain's a… ▽ More Current image de-raining methods primarily learn from a limited dataset, leading to inadequate performance in varied real-world rainy conditions. To tackle this, we introduce a new framework that enables networks to progressively expand their de-raining knowledge base by tapping into a growing pool of datasets, significantly boosting their adaptability. Drawing inspiration from the human brain's ability to continuously absorb and generalize from ongoing experiences, our approach borrow the mechanism of the complementary learning system. Specifically, we first deploy Generative Adversarial Networks (GANs) to capture and retain the unique features of new data, mirroring the hippocampus's role in learning and memory. Then, the de-raining network is trained with both existing and GAN-synthesized data, mimicking the process of hippocampal replay and interleaved learning. Furthermore, we employ knowledge distillation with the replayed data to replicate the synergy between the neocortex's activity patterns triggered by hippocampal replays and the pre-existing neocortical knowledge. This comprehensive framework empowers the de-raining network to amass knowledge from various datasets, continually enhancing its performance on previously unseen rainy scenes. Our testing on three benchmark de-raining networks confirms the framework's effectiveness. It not only facilitates continuous knowledge accumulation across six datasets but also surpasses state-of-the-art methods in generalizing to new real-world scenarios. △ Less

Submitted 3 June, 2025; originally announced June 2025.

Efficient Test-time Adaptive Object Detection via Sensitivity-Guided Pruning

Authors: Kunyu Wang, Xueyang Fu, Xin Lu, Chengjie Ge, Chengzhi Cao, Wei Zhai, Zheng-Jun Zha

Abstract: Continual test-time adaptive object detection (CTTA-OD) aims to online adapt a source pre-trained detector to ever-changing environments during inference under continuous domain shifts. Most existing CTTA-OD methods prioritize effectiveness while overlooking computational efficiency, which is crucial for resource-constrained scenarios. In this paper, we propose an efficient CTTA-OD method via prun… ▽ More Continual test-time adaptive object detection (CTTA-OD) aims to online adapt a source pre-trained detector to ever-changing environments during inference under continuous domain shifts. Most existing CTTA-OD methods prioritize effectiveness while overlooking computational efficiency, which is crucial for resource-constrained scenarios. In this paper, we propose an efficient CTTA-OD method via pruning. Our motivation stems from the observation that not all learned source features are beneficial; certain domain-sensitive feature channels can adversely affect target domain performance. Inspired by this, we introduce a sensitivity-guided channel pruning strategy that quantifies each channel based on its sensitivity to domain discrepancies at both image and instance levels. We apply weighted sparsity regularization to selectively suppress and prune these sensitive channels, focusing adaptation efforts on invariant ones. Additionally, we introduce a stochastic channel reactivation mechanism to restore pruned channels, enabling recovery of potentially useful features and mitigating the risks of early pruning. Extensive experiments on three benchmarks show that our method achieves superior adaptation performance while reducing computational overhead by 12% in FLOPs compared to the recent SOTA method. △ Less

Submitted 3 June, 2025; originally announced June 2025.

Comments: Accepted as CVPR 2025 oral paper

PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation

Authors: Kunyu Wang, Xueyang Fu, Yuanfei Bao, Chengjie Ge, Chengzhi Cao, Wei Zhai, Zheng-Jun Zha

Abstract: Continual Test-Time Adaptation (CTTA) aims to online adapt a pre-trained model to changing environments during inference. Most existing methods focus on exploiting target data, while overlooking another crucial source of information, the pre-trained weights, which encode underutilized domain-invariant priors. This paper takes the geometric attributes of pre-trained weights as a starting point, sys… ▽ More Continual Test-Time Adaptation (CTTA) aims to online adapt a pre-trained model to changing environments during inference. Most existing methods focus on exploiting target data, while overlooking another crucial source of information, the pre-trained weights, which encode underutilized domain-invariant priors. This paper takes the geometric attributes of pre-trained weights as a starting point, systematically analyzing three key components: magnitude, absolute angle, and pairwise angular structure. We find that the pairwise angular structure remains stable across diverse corrupted domains and encodes domain-invariant semantic information, suggesting it should be preserved during adaptation. Based on this insight, we propose PAID (Pairwise Angular-Invariant Decomposition), a prior-driven CTTA method that decomposes weight into magnitude and direction, and introduces a learnable orthogonal matrix via Householder reflections to globally rotate direction while preserving the pairwise angular structure. During adaptation, only the magnitudes and the orthogonal matrices are updated. PAID achieves consistent improvements over recent SOTA methods on four widely used CTTA benchmarks, demonstrating that preserving pairwise angular structure offers a simple yet effective principle for CTTA. △ Less

Submitted 3 July, 2025; v1 submitted 3 June, 2025; originally announced June 2025.

LEMONADE: A Large Multilingual Expert-Annotated Abstractive Event Dataset for the Real World

Authors: Sina J. Semnani, Pingyue Zhang, Wanyue Zhai, Haozhuo Li, Ryan Beauchamp, Trey Billing, Katayoun Kishi, Manling Li, Monica S. Lam

Abstract: This paper presents LEMONADE, a large-scale conflict event dataset comprising 39,786 events across 20 languages and 171 countries, with extensive coverage of region-specific entities. LEMONADE is based on a partially reannotated subset of the Armed Conflict Location & Event Data (ACLED), which has documented global conflict events for over a decade. To address the challenge of aggregating multil… ▽ More This paper presents LEMONADE, a large-scale conflict event dataset comprising 39,786 events across 20 languages and 171 countries, with extensive coverage of region-specific entities. LEMONADE is based on a partially reannotated subset of the Armed Conflict Location & Event Data (ACLED), which has documented global conflict events for over a decade. To address the challenge of aggregating multilingual sources for global event analysis, we introduce abstractive event extraction (AEE) and its subtask, abstractive entity linking (AEL). Unlike conventional span-based event extraction, our approach detects event arguments and entities through holistic document understanding and normalizes them across the multilingual dataset. We evaluate various large language models (LLMs) on these tasks, adapt existing zero-shot event extraction systems, and benchmark supervised models. Additionally, we introduce ZEST, a novel zero-shot retrieval-based system for AEL. Our best zero-shot system achieves an end-to-end F1 score of 58.3%, with LLMs outperforming specialized event extraction models such as GoLLIE. For entity linking, ZEST achieves an F1 score of 45.7%, significantly surpassing OneNet, a state-of-the-art zero-shot baseline that achieves only 23.7%. However, these zero-shot results lag behind the best supervised systems by 20.1% and 37.0% in the end-to-end and AEL tasks, respectively, highlighting the need for further research. △ Less

Submitted 1 June, 2025; originally announced June 2025.

Comments: Findings of ACL 2025

Shackled Dancing: A Bit-Locked Diffusion Algorithm for Lossless and Controllable Image Steganography

Authors: Tianshuo Zhang, Gao Jia, Wenzhe Zhai, Rui Yann, Xianglei Xing

Abstract: Data steganography aims to conceal information within visual content, yet existing spatial- and frequency-domain approaches suffer from trade-offs between security, capacity, and perceptual quality. Recent advances in generative models, particularly diffusion models, offer new avenues for adaptive image synthesis, but integrating precise information embedding into the generative process remains ch… ▽ More Data steganography aims to conceal information within visual content, yet existing spatial- and frequency-domain approaches suffer from trade-offs between security, capacity, and perceptual quality. Recent advances in generative models, particularly diffusion models, offer new avenues for adaptive image synthesis, but integrating precise information embedding into the generative process remains challenging. We introduce Shackled Dancing Diffusion, or SD$^2$, a plug-and-play generative steganography method that combines bit-position locking with diffusion sampling injection to enable controllable information embedding within the generative trajectory. SD$^2$ leverages the expressive power of diffusion models to synthesize diverse carrier images while maintaining full message recovery with $100\%$ accuracy. Our method achieves a favorable balance between randomness and constraint, enhancing robustness against steganalysis without compromising image fidelity. Extensive experiments show that SD$^2$ substantially outperforms prior methods in security, embedding capacity, and stability. This algorithm offers new insights into controllable generation and opens promising directions for secure visual communication. △ Less

Submitted 16 May, 2025; originally announced May 2025.

GRACE: Estimating Geometry-level 3D Human-Scene Contact from 2D Images

Authors: Chengfeng Wang, Wei Zhai, Yuhang Yang, Yang Cao, Zhengjun Zha

Abstract: Estimating the geometry level of human-scene contact aims to ground specific contact surface points at 3D human geometries, which provides a spatial prior and bridges the interaction between human and scene, supporting applications such as human behavior analysis, embodied AI, and AR/VR. To complete the task, existing approaches predominantly rely on parametric human models (e.g., SMPL), which est… ▽ More Estimating the geometry level of human-scene contact aims to ground specific contact surface points at 3D human geometries, which provides a spatial prior and bridges the interaction between human and scene, supporting applications such as human behavior analysis, embodied AI, and AR/VR. To complete the task, existing approaches predominantly rely on parametric human models (e.g., SMPL), which establish correspondences between images and contact regions through fixed SMPL vertex sequences. This actually completes the mapping from image features to an ordered sequence. However, this approach lacks consideration of geometry, limiting its generalizability in distinct human geometries. In this paper, we introduce GRACE (Geometry-level Reasoning for 3D Human-scene Contact Estimation), a new paradigm for 3D human contact estimation. GRACE incorporates a point cloud encoder-decoder architecture along with a hierarchical feature extraction and fusion module, enabling the effective integration of 3D human geometric structures with 2D interaction semantics derived from images. Guided by visual cues, GRACE establishes an implicit mapping from geometric features to the vertex space of the 3D human mesh, thereby achieving accurate modeling of contact regions. This design ensures high prediction accuracy and endows the framework with strong generalization capability across diverse human geometries. Extensive experiments on multiple benchmark datasets demonstrate that GRACE achieves state-of-the-art performance in contact estimation, with additional results further validating its robust generalization to unstructured human point clouds. △ Less

Submitted 10 May, 2025; originally announced May 2025.

On the higher moments of the error term in the Rankin-Selberg problem

Authors: Jing Huang, Yoshio Tanigawa, Wenguang Zhai, Deyu Zhang

Abstract: Let $Δ_1(x;\varphi)$ denote the error term in the classical Rankin-Selberg problem. In this paper, we consider the higher power moments of $Δ_1(x;\varphi)$ and derive the asymptotic formulas for 3-rd, 4-th and 5-th power moments, which improve the previous results. Let $Δ_1(x;\varphi)$ denote the error term in the classical Rankin-Selberg problem. In this paper, we consider the higher power moments of $Δ_1(x;\varphi)$ and derive the asymptotic formulas for 3-rd, 4-th and 5-th power moments, which improve the previous results. △ Less

Submitted 26 April, 2025; originally announced April 2025.

Comments: 14 pages

MSC Class: 11N37

Event-Based Eye Tracking. 2025 Event-based Vision Workshop

Authors: Qinyu Chen, Chang Gao, Min Liu, Daniele Perrone, Yan Ru Pei, Zuowen Wang, Zhuo Zou, Shihang Tan, Tao Han, Guorui Lu, Zhen Xu, Junyuan Ding, Ziteng Wang, Zongwei Wu, Han Han, Yuliang Wu, Jinze Chen, Wei Zhai, Yang Cao, Zheng-jun Zha, Nuwan Bandara, Thivya Kandappu, Archan Misra, Xiaopeng Lin, Hongxiang Huang , et al. (7 additional authors not shown)

Abstract: This survey serves as a review for the 2025 Event-Based Eye Tracking Challenge organized as part of the 2025 CVPR event-based vision workshop. This challenge focuses on the task of predicting the pupil center by processing event camera recorded eye movement. We review and summarize the innovative methods from teams rank the top in the challenge to advance future event-based eye tracking research.… ▽ More This survey serves as a review for the 2025 Event-Based Eye Tracking Challenge organized as part of the 2025 CVPR event-based vision workshop. This challenge focuses on the task of predicting the pupil center by processing event camera recorded eye movement. We review and summarize the innovative methods from teams rank the top in the challenge to advance future event-based eye tracking research. In each method, accuracy, model size, and number of operations are reported. In this survey, we also discuss event-based eye tracking from the perspective of hardware design. △ Less

Submitted 25 April, 2025; originally announced April 2025.