arXiv subjects

Zhe Xu

Publications and source records attributed to Zhe Xu.

At least 55 records · Page 3Linked to original sources

arXiv2026

Exploiting Low-Dimensional Manifold of Features for Few-Shot Whole Slide Image Classification

Few-shot Whole Slide Image (WSI) classification is severely hampered by overfitting. We argue that this is not merely a data-scarcity issue but a fundamentally geometric problem. Grounded in the manifold hypothesis, our analysis shows that features from pathology foundation models exhibit a low-dimensional manifold geometry that is easily perturbed by downstream models. This insight reveals a key potential issue in downstream multiple instance learning models: linear layers are geometry-agnostic and, as we show empirically, can distort the manifold geometry of the features. To address this, we propose the Manifold Residual (MR) block, a plug-and-play module that is explicitly geometry-aware. The MR block reframes the linear layer as residual learning and decouples it into two pathways: (1) a fixed, random matrix serving as a geometric anchor that approximately preserves topology while also acting as a spectral shaper to sharpen the feature spectrum; and (2) a trainable, low-rank residual pathway that acts as a residual learner for task-specific adaptation, with its structural bottleneck explicitly mirroring the low effective rank of the features. This decoupling imposes a structured inductive bias and reduces learning to a simpler residual fitting task. Through extensive experiments, we demonstrate that our approach achieves state-of-the-art results with significantly fewer parameters, offering a new paradigm for few-shot WSI classification. Code is available in https://github.com/BearCleverProud/MR-Block.

cs.CV↗

arXiv2026

Solving Football by Exploiting Equilibrium Structure of 2p0s Differential Games with One-Sided Information

For a two-player imperfect-information extensive-form game (IIEFG) with $K$ time steps and a player action space of size $U$, the game tree complexity is $U^{2K}$, causing existing IIEFG solvers to struggle with large or infinite $(U,K)$, e.g., differential games with continuous action spaces. To partially address this scalability challenge, we focus on an important class of 2p0s games where the informed player (P1) knows the payoff while the uninformed player (P2) only has a belief over the set of $I$ possible payoffs. Such games encompass a wide range of scenarios in sports, defense, cybersecurity, and finance. We prove that under mild conditions, P1's (resp. P2's) equilibrium strategy at any infostate concentrates on at most $I$ (resp. $I+1$) action prototypes. When $I\ll U$, this equilibrium structure causes the game tree complexity to collapse to $I^K$ for P1 when P2 plays best responses, and $(I+1)^K$ for P2 in a dual game where P1 plays best responses. We then show that exploiting this structure in model-free multiagent reinforcement learning and model predictive control leads to significant improvements in learning accuracy and efficiency from SOTA IIEFG solvers. Our demonstration solves a 22-player football game with continuous action spaces and $K=10$ time steps, where the offense team needs to strategically conceal their play until a critical moment in order to exploit information advantage. Code is available at https://github.com/ghimiremukesh/cams/tree/iclr

cs.GT↗

arXiv2026

Vision-R1: Incentivizing Reasoning Capability in Multimodal Large Language Models

DeepSeek-R1-Zero has successfully demonstrated the emergence of reasoning capabilities in LLMs purely through Reinforcement Learning (RL). Inspired by this breakthrough, we explore how RL can be utilized to enhance the reasoning capability of MLLMs. However, direct training with RL struggles to activate complex reasoning capabilities such as questioning and reflection in MLLMs, due to the absence of substantial high-quality multimodal reasoning data. To address this issue, we propose the reasoning MLLM, Vision-R1, to improve multimodal reasoning capability. Specifically, we first construct a high-quality multimodal CoT dataset without human annotations by leveraging an existing MLLM and DeepSeek-R1 through modality bridging and data filtering to obtain a 200K multimodal CoT dataset, Vision-R1-cold dataset. It serves as cold-start initialization data for Vision-R1. To mitigate the optimization challenges caused by overthinking after cold start, we propose Progressive Thinking Suppression Training (PTST) strategy and employ Group Relative Policy Optimization (GRPO) with the hard formatting result reward function to gradually refine the model's ability to learn correct and complex reasoning processes on a 10K multimodal math dataset. Comprehensive experiments show our model achieves an average improvement of $\sim$6% across various multimodal math reasoning benchmarks. Vision-R1-7B achieves a 73.5% accuracy on the widely used MathVista benchmark, which is only 0.4% lower than the leading reasoning model, OpenAI O1. Scaling up the amount of multimodal math data in the RL training, Vision-R1-32B and Vison-R1-72B achieves 76.4% and 78.2% MathVista benchmark scores, respectively. The datasets and code will be released in: https://github.com/Osilly/Vision-R1 .

cs.CV↗

arXiv2026

MOVA: Towards Scalable and Synchronized Video-Audio Generation

Audio is indispensable for real-world video, yet generation models have largely overlooked audio components. Current approaches to producing audio-visual content often rely on cascaded pipelines, which increase cost, accumulate errors, and degrade overall quality. While systems such as Veo 3 and Sora 2 emphasize the value of simultaneous generation, joint multimodal modeling introduces unique challenges in architecture, data, and training. Moreover, the closed-source nature of existing systems limits progress in the field. In this work, we introduce MOVA (MOSS Video and Audio), an open-source model capable of generating high-quality, synchronized audio-visual content, including realistic lip-synced speech, environment-aware sound effects, and content-aligned music. MOVA employs a Mixture-of-Experts (MoE) architecture, with a total of 32B parameters, of which 18B are active during inference. It supports IT2VA (Image-Text to Video-Audio) generation task. By releasing the model weights and code, we aim to advance research and foster a vibrant community of creators. The released codebase features comprehensive support for efficient inference, LoRA fine-tuning, and prompt enhancement.

cs.CV↗

arXiv2026

StepScorer: Accelerating Reinforcement Learning with Step-wise Scoring and Psychological Regret Modeling

Reinforcement learning algorithms often suffer from slow convergence due to sparse reward signals, particularly in complex environments where feedback is delayed or infrequent. This paper introduces the Psychological Regret Model (PRM), a novel approach that accelerates learning by incorporating regret-based feedback signals after each decision step. Rather than waiting for terminal rewards, PRM computes a regret signal based on the difference between the expected value of the optimal action and the value of the action taken in each state. This transforms sparse rewards into dense feedback signals through a step-wise scoring framework, enabling faster convergence. We demonstrate that PRM achieves stable performance approximately 36\% faster than traditional Proximal Policy Optimization (PPO) in benchmark environments such as Lunar Lander. Our results indicate that PRM is particularly effective in continuous control tasks and environments with delayed feedback, making it suitable for real-world applications such as robotics, finance, and adaptive education where rapid policy adaptation is critical. The approach formalizes human-inspired counterfactual thinking as a computable regret signal, bridging behavioral economics and reinforcement learning.

cs.LG↗

arXiv2026

Global Feature Enhancing and Fusion Framework for Strain Gauge Time Series Classification

Strain Gauge Status (SGS) time series recognition is crucial in the field of intelligent manufacturing based on the Internet of Things, as accurate identification helps timely detection of failed mechanical components, avoiding accidents. The loading and unloading sequences generated by strain gauges can be identified through time series classification (TSC) algorithms. Recently, deep learning models, e.g., convolutional neural networks (CNNs) have shown remarkable success in the TSC task, as they can extract discriminative local features from the subsequences to identify the time series. However, we observe that only the local features may not be sufficient for expressing the time series, especially when the local sub-sequences between different time series are very similar, e.g., SGS data of aircraft wings in static strength experiments. Nevertheless, CNNs suffer from the limitation in extracting global features due to the nature of convolution operations. For extracting global features to more comprehensively represent the SGS time series, we propose two insights: (i) Constructing global features through feature engineering. (ii) Learning high-order relationships between local features to capture global features. To realize and utilize them, we propose a hypergraph-based global feature learning and fusion framework, which learns and fuses global features for semantic consistency to enhance the representation of SGS time series, thereby improving recognition accuracy. Our method designs are validated on industrial SGS and public UCR datasets, showing better generalization for unseen data in SGS recognition. The code is available at the link https://github.com/Meteor-Stars/GFEF.

cs.LG↗

arXiv2026

How to Make LMs Strong Node Classifiers?

Language Models (LMs) are increasingly challenging the dominance of domain-specific models, such as Graph Neural Networks (GNNs) and Graph Transformers (GTs), in graph learning tasks. Following this trend, we propose a novel approach that empowers off-the-shelf LMs to achieve performance comparable to state-of-the-art (SOTA) GNNs on node classification tasks, without requiring any architectural modification. By preserving the LM's original architecture, our approach retains a key benefit of LM instruction tuning: the ability to jointly train on diverse datasets, fostering greater flexibility and efficiency. To achieve this, we introduce two key augmentation strategies: (1) Enriching LMs' input using topological and semantic retrieval methods, which provide richer contextual information, and (2) guiding the LMs' classification process through a lightweight GNN classifier that effectively prunes class candidates. Our experiments on real-world datasets show that backbone Flan-T5 LMs equipped with these augmentation strategies outperform SOTA text-output node classifiers and are comparable to top-performing vector-output node classifiers. By bridging the gap between specialized node classifiers and general LMs, this work paves the way for more versatile and widely applicable graph learning models. We will open-source the code upon publication.

cs.CL↗

arXiv2026

Inferring Causal Graph Temporal Logic Formulas to Expedite Reinforcement Learning in Temporally Extended Tasks

Decision-making tasks often unfold on graphs with spatial-temporal dynamics. Black-box reinforcement learning often overlooks how local changes spread through network structure, limiting sample efficiency and interpretability. We present GTL-CIRL, a closed-loop framework that simultaneously learns policies and mines Causal Graph Temporal Logic (Causal GTL) specifications. The method shapes rewards with robustness, collects counterexamples when effects fail, and uses Gaussian Process (GP) driven Bayesian optimization to refine parameterized cause templates. The GP models capture spatial and temporal correlations in the system dynamics, enabling efficient exploration of complex parameter spaces. Case studies in gene and power networks show faster learning and clearer, verifiable behavior compared to standard RL baselines.

cs.AI↗

arXiv2025

Jupiter: Enhancing LLM Data Analysis Capabilities via Notebook and Inference-Time Value-Guided Search

Large language models (LLMs) have shown great promise in automating data science workflows, but existing models still struggle with multi-step reasoning and tool use, which limits their effectiveness on complex data analysis tasks. To address this, we propose a scalable pipeline that extracts high-quality, tool-based data analysis tasks and their executable multi-step solutions from real-world Jupyter notebooks and associated data files. Using this pipeline, we introduce NbQA, a large-scale dataset of standardized task-solution pairs that reflect authentic tool-use patterns in practical data science scenarios. To further enhance multi-step reasoning, we present Jupiter, a framework that formulates data analysis as a search problem and applies Monte Carlo Tree Search (MCTS) to generate diverse solution trajectories for value model learning. During inference, Jupiter combines the value model and node visit counts to efficiently collect executable multi-step plans with minimal search steps. Experimental results show that Qwen2.5-7B and 14B-Instruct models on NbQA solve 77.82% and 86.38% of tasks on InfiAgent-DABench, respectively-matching or surpassing GPT-4o and advanced agent frameworks. Further evaluations demonstrate improved generalization and stronger tool-use reasoning across diverse multi-step reasoning tasks. Code and data are available at https://github.com/microsoft/Jupiter.

cs.AI↗

arXiv2025

Multiwavelength Observations for a Double-decker Filament Channel in AR 13102

We present the observational evidence of the existence of a double-decker filament channel (FC) by using observations in extreme ultraviolet and Hα wavelengths. For both FCs, the east foot-point roots in the active region (AR), while the west one roots in the remote quiet region. The bottom FC (FC1) appears as intermittent filaments. Within the AR, the FC1 appears as an S-shaped filament (F1), which consisted of two J-shaped filaments (F1S/F1N for the south/north one). For the upper one (FC2), only the east part is filled with dark plasma and visible as a small filament (F2). Its east foot-point roots around the junction of F1S and F1N. Initially, due to the recurrent reconnections, F1N and F1S link to each other and form a new filament (F3) thread by thread. Meanwhile, the heated plasma, which appears as brightening features, flows from the east foot-point of F2 to the west, and becomes invisible about 1.1$\times$10^{5} km away. The failed eruption of F1S is triggered by the reconnection, which appears as the brightening threads changing their configuration from crossed to quasiparallel in between the F1S and F3, and is confined by the upper magnetic field. Associated with the eruption, the distant invisible plasma becomes visible as a brightening feature. It continuously flows to the remote foot-point, and becomes invisible before reaching it. The brightening plasma flow outlines the skeleton of FC2 gradually. The observations show the existence of a double-decker FC, as a magnetic structure, before they appear as a brightening/dark feature when fully filled with hot/cool plasma.

astro-ph.SR↗

arXiv2025

Causal-HalBench: Uncovering LVLMs Object Hallucinations Through Causal Intervention

Large Vision-Language Models (LVLMs) often suffer from object hallucination, making erroneous judgments about the presence of objects in images. We propose this primar- ily stems from spurious correlations arising when models strongly associate highly co-occurring objects during train- ing, leading to hallucinated objects influenced by visual con- text. Current benchmarks mainly focus on hallucination de- tection but lack a formal characterization and quantitative evaluation of spurious correlations in LVLMs. To address this, we introduce causal analysis into the object recognition scenario of LVLMs, establishing a Structural Causal Model (SCM). Utilizing the language of causality, we formally de- fine spurious correlations arising from co-occurrence bias. To quantify the influence induced by these spurious correla- tions, we develop Causal-HalBench, a benchmark specifically constructed with counterfactual samples and integrated with comprehensive causal metrics designed to assess model ro- bustness against spurious correlations. Concurrently, we pro- pose an extensible pipeline for the construction of these coun- terfactual samples, leveraging the capabilities of proprietary LVLMs and Text-to-Image (T2I) models for their genera- tion. Our evaluations on mainstream LVLMs using Causal- HalBench demonstrate these models exhibit susceptibility to spurious correlations, albeit to varying extents.

cs.AI↗

arXiv2025

RedOne 2.0: Rethinking Domain-specific LLM Post-Training in Social Networking Services

As a key medium for human interaction and information exchange, social networking services (SNS) pose unique challenges for large language models (LLMs): heterogeneous workloads, fast-shifting norms and slang, and multilingual, culturally diverse corpora that induce sharp distribution shift. Supervised fine-tuning (SFT) can specialize models but often triggers a ``seesaw'' between in-distribution gains and out-of-distribution robustness, especially for smaller models. To address these challenges, we introduce RedOne 2.0, an SNS-oriented LLM trained with a progressive, RL-prioritized post-training paradigm designed for rapid and stable adaptation. The pipeline consist in three stages: (1) Exploratory Learning on curated SNS corpora to establish initial alignment and identify systematic weaknesses; (2) Targeted Fine-Tuning that selectively applies SFT to the diagnosed gaps while mixing a small fraction of general data to mitigate forgetting; and (3) Refinement Learning that re-applies RL with SNS-centric signals to consolidate improvements and harmonize trade-offs across tasks. Across various tasks spanning three categories, our 4B scale model delivers an average improvements about 2.41 over the 7B sub-optimal baseline. Additionally, RedOne 2.0 achieves average performance lift about 8.74 from the base model with less than half the data required by SFT-centric method RedOne, evidencing superior data efficiency and stability at compact scales. Overall, RedOne 2.0 establishes a competitive, cost-effective baseline for domain-specific LLMs in SNS scenario, advancing capability without sacrificing robustness.

cs.AI↗

arXiv2025

MindGYM: What Matters in Question Synthesis for Thinking-Centric Fine-Tuning?

Large foundation models face challenges in acquiring transferable, structured thinking abilities, especially when supervised with rigid templates or crowd-annotated instruction datasets. Unlike prior approaches, we focus on a thinking-centric data synthesis paradigm that enables models to evolve through self-generated, cognitively guided data. We propose MindGYM, a structured and scalable framework for question synthesis, composed of: (1) Cognitive Thinking Process Injection, which infuses high-level reasoning objectives to shape the model's synthesis behavior; (2) Seed Single-Hop Question Synthesis, generating atomic questions from diverse semantic types to encourage broader thinking; and (3) Challenging Multi-Hop QA Synthesis, composing more complex multi-hop questions based on QA seeds for deeper reasoning. Detailed analysis shows that synthetic data generated by our method achieves 16.7% higher average quality and 67.91% lower quality variance compared to baseline sources, highlighting that both high-quality and self-contained data are essential for effective, thinking-oriented fine-tuning. MindGYM improves performance on six reasoning benchmarks, achieving gains of up to 16% on MathVision using only 400 data samples, and generalizable improvements across different model sizes and architectures. MindGYM underscores the viability of self-challenging mechanisms in refining large model capabilities while minimizing human intervention and resource demands. Code and data are released to promote data-centric research into self-evolving foundation models driven by their internal reasoning capabilities.

cs.CV↗

arXiv2025

Magnetic Field Configuration of a Quiescent Prominence Revealed by Large-amplitude Longitudinal Oscillations in End-view Observations

Prominence seismology, applied to the large-amplitude longitudinal oscillation, is used to indirectly diagnose the geometry and strength of the magnetic fields inside the prominence. In this paper, combining imaging and spectroscopic data, the magnetic field configuration of a quiescent prominence is revealed by large-amplitude longitudinal oscillations observed in end view on 2023 December 4. Particularly, the prominence oscillation involved blueshift velocities in Dopplergrams and horizontal motions in extreme-ultraviolet (EUV) images. Originally, the prominence oscillation was triggered by the collision and heating of an adjoining hot structure associated with two coronal jets. The oscillation involved two groups of signals with similar oscillatory parameters, a three-dimensional (3D) initial amplitude of 40 Mm and a 3D velocity amplitude of 48 km/s, both lasting for 4 cycles with a period of 77 minutes, with a phase difference of pi/4. While the angle between 3D velocities and the prominence axis ranges from 10 to 30. Two methods, utilizing time-distance diagrams and velocity fields, are employed to calculate the curvature radius of magnetic dips supporting the prominence materials. Both methods yield similar value ranges and trends from the bottom to the top of magnetic dips, with the curvature radius increasing from 90 Mm to 220 Mm, then decreasing to 10 Mm, with transverse magnetic field strength 25 Gauss. From this, the realistic 3D geometry of the prominence magnetic dips is determined to be sinusoidal. To the best of our knowledge, we present the first accurate calculation of the 3D curvature radius and geometry of the prominence magnetic dips based on longitudinal oscillatory motions.

astro-ph.SR↗

arXiv2025

Expediting Reinforcement Learning by Incorporating Knowledge About Temporal Causality in the Environment

Reinforcement learning (RL) algorithms struggle with learning optimal policies for tasks where reward feedback is sparse and depends on a complex sequence of events in the environment. Probabilistic reward machines (PRMs) are finite-state formalisms that can capture temporal dependencies in the reward signal, along with nondeterministic task outcomes. While special RL algorithms can exploit this finite-state structure to expedite learning, PRMs remain difficult to modify and design by hand. This hinders the already difficult tasks of utilizing high-level causal knowledge about the environment, and transferring the reward formalism into a new domain with a different causal structure. This paper proposes a novel method to incorporate causal information in the form of Temporal Logic-based Causal Diagrams into the reward formalism, thereby expediting policy learning and aiding the transfer of task specifications to new environments. Furthermore, we provide a theoretical result about convergence to optimal policy for our method, and demonstrate its strengths empirically.

cs.LG↗

arXiv2025

Decentralizing Multi-Agent Reinforcement Learning with Temporal Causal Information

Reinforcement learning (RL) algorithms can find an optimal policy for a single agent to accomplish a particular task. However, many real-world problems require multiple agents to collaborate in order to achieve a common goal. For example, a robot executing a task in a warehouse may require the assistance of a drone to retrieve items from high shelves. In Decentralized Multi-Agent RL (DMARL), agents learn independently and then combine their policies at execution time, but often must satisfy constraints on compatibility of local policies to ensure that they can achieve the global task when combined. In this paper, we study how providing high-level symbolic knowledge to agents can help address unique challenges of this setting, such as privacy constraints, communication limitations, and performance concerns. In particular, we extend the formal tools used to check the compatibility of local policies with the team task, making decentralized training with theoretical guarantees usable in more scenarios. Furthermore, we empirically demonstrate that symbolic knowledge about the temporal evolution of events in the environment can significantly expedite the learning process in DMARL.

cs.LG↗

arXiv2025

Observational study of chromospheric jets in and around a sunspot observed by NVST and SDO

To better understand the characteristics, driving mechanisms, and potential heating contributions of chromospheric jets, we analyze two contrasting types: one originating from within the sunspot penumbra (inside jets), and the other originating from outside the penumbra (outside jets). Statistical analysis of 100 jets (50 inside jets and 50 outside jets) reveals that inside jets have a projected velocity range of 4--14~km\,s$^{-1}$, a length range of 1--4~Mm, a width range of 0.2--0.6~Mm, and a lifetime range of 135--450~s, with mean values of 7.90~km\,s$^{-1}$, 2.61~Mm, 0.41~Mm, and 260~s, respectively. About 52\% of inside jets are associated with brightenings in H$α$ blue wing images, and some show high-temperature signatures, suggesting a connection with localized energy release. In contrast, outside jets have higher velocities (8--50~km\,s$^{-1}$, average 19.04~km\,s$^{-1}$), greater lengths (average 6.26~Mm, up to 27.27~Mm), slightly larger widths (average 0.46~Mm), and longer lifetimes (135--630~s, average 327~s). They typically originate from regions of opposite magnetic polarities and are associated with magnetic flux emergence and EUV brightenings. Some outside jets correspond to coronal jets with inverted Y-shaped structures and temperatures exceeding one million Kelvin. Our results suggest that both jet types are driven by magnetic reconnection occurring in distinct magnetic field configurations and contribute to chromospheric and coronal heating.

astro-ph.SR↗

arXiv2025

RedOne: Revealing Domain-specific LLM Post-Training in Social Networking Services

As a primary medium for modern information dissemination, social networking services (SNS) have experienced rapid growth, which has proposed significant challenges for platform content management and interaction quality improvement. Recently, the development of large language models (LLMs) has offered potential solutions but existing studies focus on isolated tasks, which not only encounter diminishing benefit from the data scaling within individual scenarios but also fail to flexibly adapt to diverse real-world context. To address these challenges, we introduce RedOne, a domain-specific LLM designed to break the performance bottleneck of single-task baselines and establish a comprehensive foundation for the SNS. RedOne was developed through a three-stage training strategy consisting of continue pretraining, supervised fine-tuning, and preference optimization, using a large-scale real-world dataset. Through extensive experiments, RedOne maintains strong general capabilities, and achieves an average improvement up to 14.02% across 8 major SNS tasks and 7.56% in SNS bilingual evaluation benchmark, compared with base models. Furthermore, through online testing, RedOne reduced the exposure rate in harmful content detection by 11.23% and improved the click page rate in post-view search by 14.95% compared with single-tasks finetuned baseline models. These results establish RedOne as a robust domain-specific LLM for SNS, demonstrating excellent generalization across various tasks and promising applicability in real-world scenarios.

cs.LG↗