Searcharxiv⌕ Search

arXiv subjects

Jian Zhang

Publications and source records attributed to Jian Zhang.

At least 55 records · Page 3Linked to original sources
arXiv2026

CausalOPD: First-Wrong-Step Supervision for Distilling Causal Chain Reasoning

Many critical reasoning tasks, including clinical diagnosis, legal judgment, and industrial fault diagnosis, require step-dependent causal chains in which early errors propagate and correct conclusions can mask invalid reasoning. Although large language models perform well on such tasks, privacy, latency, and controllability motivate distillation into locally deployable models. Standard trajectory imitation does not correct process errors on the student's own rollout distribution. We propose CausalOPD, a curriculum online process distillation framework. A knowledge-augmented teacher first provides trajectories grounded in domain-specific causal rules, entity relations, and structural constraints. The student then generates on-policy trajectories, and the teacher identifies the first wrong step, defined as the earliest transition that verifiably violates available constraints. Starting from the verified prefix, short-horizon reinforcement learning repairs this localized failure. A causal-stage curriculum advances from evidence-level to mechanism-level and conclusion-level errors, following their propagation order. Across three domains, CausalOPD improves average path correctness by 23.4 percentage points over sequence-level online process distillation and reduces the right-label-wrong-reasoning rate from 15.7% to 4.4%. The domain-specific 8B students also surpass both evaluated proprietary references in path correctness across all domains.

cs.LG↗
arXiv2026

DiagLoop: A Counterfactual Data Flywheel with Stage-Localized Reinforcement for Diagnostic LLMs

Causal diagnostic models must explain how conclusions follow from evidence because diagnoses guide repairs and treatments. Yet serious cases are scarce, records rarely contain reasoning paths, and data transfer poorly across configurations, complicating local deployment. We present DiagLoop, a counterfactual data flywheel that converts codified physical relations or clinical guidelines, authored once per mechanism family, into training supervision beyond recorded cases. A training-only teacher proposes counterfactual worlds by varying causes, contexts, and observations, while an independent hybrid checker admits only valid worlds. The student reasons through symptom abstraction, causal-chain construction, and root-cause attribution. Stage-specific criteria identify its earliest failure. For nonterminal failures, a bounded repair probes downstream competence, and the resulting weakness profile guides subsequent data generation. Stage-localized reinforcement learning updates only the model-generated continuation, while replay and preservation reduce forgetting. The same criteria govern admission, attribution, reward, and regeneration through checks separate from the proposer. Using only synthesized scenarios and no case-level expert reasoning annotations, the resulting 8B model improves strict path correctness over the strongest conventional baseline. Gains are 11.6 points across eight industrial systems and 5.5 points across ten disease categories. Gains over a deranged-routing control are 3.9 and 2.3 points, respectively. The model also exceeds the evaluated proprietary references in both domains, even when they receive few-shot examples or the specification in context.

cs.LG↗
arXiv2026

AgenticVAU: Multi-Agent Explore-Verify Reasoning for Video Anomaly Understanding

Video anomaly understanding (VAU) focuses on comprehensively interpreting abnormal events in videos, requiring models to identify anomalous occurrences, discover their supporting evidence, and explain the underlying causes beyond simple anomaly detection. Existing VAU methods often rely on specialized training or limited observations, restricting generalization or evidence coverage. Although single-agent alternatives support adaptive video observation, they still integrate exploration, observation, and decision-making within a unified reasoning process, offering limited role specialization and structured evidence coordination. To address these limitations, we present AgenticVAU, a training-free multi-agent framework that casts VAU as an explore--verify process, where the system first discovers potential anomalies and then verifies them through targeted observations. To achieve this, four specialized agents are introduced to handle visual-rule construction, search planning, video observation, and final decision, respectively. These agents communicate through an anchor registry, a shared evidence memory that binds each observation. Guided by this agent framework, AgenticVAU interleaves broad temporal exploration, dense local verification, and cross-interval comparison until sufficient evidence is collected. We conduct extensive experiments on the ECVA, UCF-Crime, and MSAD subsets of VAU-Bench, the results show that AgenticVAU outperforms zero-shot inference and reinforcement learning-based baselines, demonstrating the value of multi-agent collaboration for video anomaly understanding.

cs.CV↗
arXiv2026

Chart Specification: Structural Representations for Incentivizing VLM Reasoning in Chart-to-Code Generation

Vision-Language Models (VLMs) have shown promise in generating plotting code from chart images, yet achieving structural fidelity remains challenging. Existing approaches largely rely on supervised fine-tuning, encouraging surface-level token imitation rather than faithful modeling of underlying chart structure, which often leads to hallucinated or semantically inconsistent outputs. We propose Chart Specification, a structured intermediate representation that shifts training from text imitation to semantically grounded supervision. Chart Specification filters syntactic noise to construct a structurally balanced training set and supports a Spec-Align Reward that provides fine-grained, verifiable feedback on structural correctness, enabling reinforcement learning to enforce consistent plotting logic. Experiments on three public benchmarks show that our method consistently outperforms prior approaches. With only 3K training samples, we achieve strong data efficiency, surpassing leading baselines by up to 61.7% on complex benchmarks, and scaling to 4K samples establishes new state-of-the-art results across all evaluated metrics. Overall, our results demonstrate that precise structural supervision offers an efficient pathway to high-fidelity chart-to-code generation. Code and dataset are available at: https://github.com/Mighten/chart-specification-paper

cs.CV↗
arXiv2026

Piggybacking on Perception: Stealthy Concurrent Audio Prompt Injections against Multimodal LLM Agents

Large Language Model (LLM)-driven multimodal agents are increasingly deployed to execute autonomous tasks via continuous audio interaction. While this paradigm enhances interaction naturalness, it introduces a critical yet under-explored attack surface, as audio inputs inevitably contain environmental noise beyond user control. In this paper, we investigate concurrent audio prompt injection attacks targeting multimodal agents. Distinct from traditional acoustic attacks on voice devices, we propose novel techniques for instruction augmentation and scenario concealment. These methods allow malicious audio instructions to imperceptibly "piggyback" onto user speech, thereby hijacking agents to execute malicious actions. To systematically quantify this threat, we construct AudioAgentSecurity, the first comprehensive benchmark for audio instruction injection attacks, encompassing 8 real-world task scenarios and 10 distinct attack patterns. We evaluate 11 state-of-the-art agents, including Gemini 3 Pro and GPT-4o-audio. Notably, our methods achieve an average Attack Success Rate (ASR) of 69.10\% against the advanced Gemini 3 Pro. To counter this threat, we further introduce Cascaded Audio Decoupling and Verification (CADV), a defense mechanism based on source separation and consistency analysis. Compared with existing prompt-level defenses, CADV achieving up to 96\% detection accuracy and providing effective protection against a broad range of acoustic injection attacks. Finally, real-world experiments with human volunteers on Doubao AI Smartphone in diverse dynamic real-world scenarios confirm the attacks' high stealth and efficacy, while demonstrating that our defense reliably mitigates these vulnerabilities.

cs.CR↗
arXiv2026

The MiniMax-M2 Series: Mini Activations Unleashing Max Real-World Intelligence

We introduce the MiniMax-M2 series, a family of Mixture-of-Experts language models built around the principle that mini activations can unleash maximum real-world intelligence. The flagship M2 contains 229.9B total parameters with only 9.8B activated per token. Designed end-to-end for agentic deployment, the M2 series rests on three components: (i) agent-driven data pipelines producing large-scale, verifiable trajectories across agentic coding and agentic cowork, each grounded in an executable workspace and an artifact-aligned reward; (ii) Forge, a scalable agent-native RL system that adapts to long-horizon agent trajectories, paired with windowed-FIFO scheduling, prefix-tree merging, inference optimization, and a clean training-inference-agent decoupling that supports both white-box and black-box agents; (iii) the latest M2.7 checkpoint takes an early step toward self-evolution -- autonomously debugging training runs and modifying its own scaffold. Across M2 through M2.7, this combination translates a mini-activation footprint into frontier-tier performance on agentic coding, deep search, office-task, and reasoning benchmarks.

cs.AI↗
arXiv2026

IMFuse: Instance-Aware Multi-Layer Fusion for LLM-Enhanced Sequential Recommendation

Recent advancements in Large Language Models (LLMs) have significantly enhanced sequential recommendation by encoding rich item textual information into semantic representations. However, existing methods typically rely on the final-layer hidden states of LLMs, overlooking potentially useful semantic signals encoded in other layers. Through empirical analysis, we reveal the limitations of this practice: final-layer representations often suffer from dimensional collapse, whereas intermediate layers preserve complementary, coarse-to-fine semantic knowledge. Furthermore, we observe that different items exhibit heterogeneous layer-wise representation evolution, making a uniform layer selection sub-optimal. To bridge this gap, we propose IMFuse, an instance-aware multi-layer fusion strategy designed for LLM-enhanced recommendation. Instead of relying on a single layer, IMFuse adaptively aggregates multi-layer semantic information by learning global dimension-wise layer preferences to capture general semantic contributions. To address item-level heterogeneity, IMFuse introduces an instance-aware expert modulation mechanism that dynamically adjusts these global preferences, generating personalized, item-specific semantic representations. Extensive experiments across four real-world datasets demonstrate the effectiveness of IMFuse. It consistently outperforms state-of-the-art baselines with an average relative improvement of 6.72%, while introducing limited parameter and computational overhead.

cs.IR↗
arXiv2026

Irreducibility of the tensor product of Yangian \( \mathrm{Y}(\mathfrak{gl}_{m|n}) \) evaluation modules

The evaluation homomorphism from the super Yangian \( \mathrm{Y}(\mathfrak{gl}_{m|n}) \) to \( \mathrm{U}(\mathfrak{gl}_{m|n}) \) induces a \( \mathrm{Y}(\mathfrak{gl}_{m|n}) \)-module structure on any finite dimensional simple \( \mathrm{U}(\mathfrak{gl}_{m|n}) \)-module \( L(λ) \). In this paper, we give necessary and sufficient conditions for the tensor product of such evaluation \( \mathrm{Y}(\mathfrak{gl}_{m|n}) \)-modules, \( L_g(λ) \otimes L_h(γ) \), to be simple, provided each of \( λ\) and \( γ\) is either covariant tensor or essentially typical. Our proof is based on the existence of a Gelfand--Tsetlin basis for finite dimensional simple \( \mathrm{U}(\mathfrak{gl}_{m|n}) \)-modules with highest weights that belong to these two families: covariant tensor and essentially typical. The obtained result is a super analogue of the Molev's result for the classical Yangian \( \mathrm{Y}(\mathfrak{gl}_n) \). Combining this with the binary property of tensor products of covariant evaluation modules, we obtain an irreducibility criterion for arbitrary tensor products of covariant evaluation modules.

math.RT↗
arXiv2026

PC-Edit: Prompt-Contrastive Region Discovery and Region-Guided Editing

Replacing an object with one that differs in category or shape requires complete source removal, natural target formation unconstrained by the source silhouette, and preservation of unrelated content. Existing training-free editors either localize edits from terminal predictions under source and target prompts or preserve unrelated content through spatially unselective source-feature reuse without explicit region discovery. Before reaching the terminal predictions, prompt-induced semantic differences undergo additional network transformations that may obscure their spatial localization, reducing localization precision. Spatially unselective feature reuse forces a trade-off between edit completeness and background preservation. Therefore, we propose PC-Edit, a prompt-contrastive framework for training-free MM-DiT editing. PC-Edit contrasts the image-token attention outputs under the source and target prompts, capturing prompt-induced semantic differences directly where text-conditioned information is delivered to image tokens. The same contrast identifies a source-erasure region during inversion and a target-emergence region during denoising. Their union suppresses source remnants while allowing the target object to form naturally. PC-Edit further couples region discovery and background preservation within each sampling step by estimating the current edit region from preceding attention blocks and immediately injecting cached source K/V features outside it in subsequent blocks, thereby protecting unrelated content before the latent update. Experiments on PIE-Bench and our EditRegion-Bench, with human-verified edit-region annotations for single- and multi-object addition and replacement, show that PC-Edit achieves the best editing quality and background preservation among methods without user-specified edit regions.

cs.CV↗
arXiv2026

Invariants of the extended twisted $h$-Yangian

We investigate the extended twisted $h$-Yangian ${\rm Y}_{N,h}^{\text{tw}}$, a certain algebra which admits both the orthogonal and the symplectic $h$-Yangian as its quotients. We show that ${\rm Y}_{N,h}^{\text{tw}}$ is naturally equipped with the structure of restricted module for a certain algebra ${\rm DY}_{N,h,c}^{tw}$, which resembles the quantum double, as well as with the structure of $ϕ$-coordinated quasi module for the Etingof-Kazhdan quantum affine vertex algebra associated with the trigonometric $R$-matrix of type $A$. Finally, we demonstrate how the elements of the quantum Feigin--Frenkel center give rise to explicit formulas for central elements of ${\rm DY}_{N,h,c}^{tw}$ and invariants of ${\rm Y}_{N,h}^{\text{tw}}$ at the critical level, as well as to commutative families in the orthogonal and symplectic $h$-Yangians.

math.QA↗
arXiv2026

Workflow-GYM: Towards Long-Horizon Evaluation of Computer-use Agentic tasks in Real-World Professional Fields

Recent years have witnessed the rapid evolution of AI agents toward handling increasingly complex, real-world tasks. However, existing benchmarks rarely evaluate whether agents can operate graphical user interfaces to complete long-horizon, high-value professional workflows across diverse domains. Current GUI benchmarks still predominantly focus on general-purpose software, relatively simple applications, and short-horizon tasks, leaving it largely unknown whether modern agents can follow user instructions to autonomously operate domain-specific professional software and accomplish economically valuable work in an end-to-end manner. To bridge this gap, we introduce Workflow-GYM, a benchmark for long-horizon GUI tasks centered on professional domains and specialized software environments. Through extensive experiments on state-of-the-art models, we find that even the strongest models achieve only slightly above 30% success rates, highlighting that professional long-horizon GUI workflows remain highly challenging for current GUI agents. Further analysis reveals that current agents struggle to maintain long-horizon workflow consistency, frequently exhibiting workflow stage omission, error propagation, objective drift, and insufficient understanding of professional software environments. Our findings provide important insights into the limitations of current agent systems and suggest key directions for the next generation of GUI-agent research.

cs.AI↗
arXiv2026

FuncDroid: Towards Inter-Functional Flows for Comprehensive Mobile App GUI Testing

As mobile application (app) functionalities grow increasingly complex and their iterations accelerate, ensuring high reliability presents significant challenges. While functionality-oriented GUI testing has attracted growing research attention, existing approaches largely overlook interactions across functionalities, making them ineffective at uncovering deep bugs hidden in inter-functional behaviors. To fill this gap, we first design a Functional Flow Graph (FFG), a behavioral model that explicitly captures an app's functional units and their inter-functional interactions. Based on the FFG, we further introduce an inter-functional-flow-oriented GUI testing approach with the dual goals of precise model construction and deep bug detection. This approach is realized through a long-short-term-view-guided testing process. By combining two complementary test-generation views, it can adaptively refine functional boundaries and systematically explore inter-functional flows under diverse triggering conditions. We implement our approach in a tool called FuncDroid, and evaluate it on two benchmarks: (1) a widely-used open-source benchmark with 50 reproducible crash bugs and (2) a diverse set of 52 popular commercial apps. Experimental results demonstrate that FuncDroid significantly outperforms state-of-the-art baselines in both coverage (+28%) and bug detection number (+107%). Moreover, FuncDroid successfully uncovers 18 previously unknown non-crash functional bugs in commercial apps, confirming its practical effectiveness.

cs.SE↗
arXiv2026

Motif-Mamba: network motif improved mamba for long-range sequence modeling

Efficient long-sequence modeling remains a central challenge for large language models, as self-attention scales quadratically with sequence length. Mamba offers a linear-time alternative through selective state space recurrence, but its predominantly diagonal state transitions restrict explicit interactions among state dimensions. We propose Motif-Mamba, a structured state space model that augments Mamba with a motif-constrained low-rank recurrent pathway. Inspired by the dynamics of three-node network motifs, the proposed pathway projects hidden states into a compact dynamical subspace, imposes motif-guided interactions, and maps the resulting dynamics back to the original state space. This design enhances cross-dimensional communication while preserving the linear-time recurrent structure of Mamba. Experiments on long-sequence extrapolation, language modeling benchmarks, and brain--computer interface decoding show consistent improvements over Mamba backbones, suggesting that motif-guided low-rank dynamics provide an effective structural prior for long-range sequence modeling.

cs.AI↗
arXiv2026

Central Elements and Determinantal Identities in the Elliptic Quantum Algebra \( \mathcal{A}_{q,p}(\widehat{\mathfrak{gl}}_N)\)

Elliptic quantum algebra is the algebraic structure characterized by the elliptic solution of the Yang-Baxter equation. In this paper, we construct a family of central elements \( \mathfrak{z}(z) \) for the elliptic quantum algebra \(\mathcal{A}_{q,p}(\widehat{\mathfrak{gl}}_{N})\) and show that they can be expressed as quantum determinants, yielding an elliptic analogue of the Liouville formula. In addition, we establish determinantal identities, including Jacobi's ratio theorem and Sylvester's theorem.

math.QA↗
arXiv2026

LLM-Enhanced Hierarchical Heterogeneous Graph Representation Learning for Malicious Python Package Detection

Malicious Python packages have become a major threat to software supply chain ecosystems due to the widespread adoption of open-source repositories such as PyPI. Existing learning-based detection methods struggle to capture the hierarchical organization and heterogeneous interactions among different program entities. Although Large Language Models (LLMs) have demonstrated strong capabilities in code understanding and semantic reasoning, they are rarely integrated with structural program representations for fine-grained malicious behavior analysis. In this paper, we propose an LLM-enhanced hierarchical heterogeneous graph representation learning framework for malicious Python package detection. The framework constructs a hierarchical heterogeneous code graph that explicitly models heterogeneous code entities and different types of structural dependencies. LLMs are further leveraged to infer function-level semantic roles, introducing an additional layer of semantic heterogeneity. Based on this graph, we develop a hierarchical heterogeneous graph neural network that performs type-aware message passing over different node and edge categories, effectively modeling malicious behavior propagation for accurate package-level classification. The framework also incorporates a function-level attribution mechanism which, combined with LLM reasoning, automatically identifies suspicious functions and localizes fine-grained malicious behaviors without human expert intervention. Extensive experiments on real-world datasets show that our framework consistently outperforms traditional machine learning methods, graph-based detectors, and state-of-the-art LLMs across packages with varying sizes and dependency complexities, while providing accurate, robust, and interpretable malicious behavior localization.

cs.CR↗
arXiv2026

KV-RM: Regularizing KV-Cache Movement for Static-Graph LLM Serving

Static-graph LLM decoders provide predictable launches, fixed tensor shapes, and low submission overhead, but online decoding exposes highly irregular KV-cache behavior: request lengths differ, EOS events arrive asynchronously, and logical histories fragment over time. Dynamic runtimes recover flexibility through paged KV management and step-level scheduling, while static-graph executors often over-reserve memory and suffer burst-time latency outliers. This paper studies whether much of this variability can be absorbed below a fixed decode interface. We present KV-RM, a runtime design that regularizes KV-cache movement beneath a static-graph LLM decoder. KV-RM decouples logical KV histories from physical storage, tracks active KV state through a block pager, and materializes each decode step through a single committed descriptor. A merge-staged transport path coalesces non-contiguous KV mappings into a small number of large transfer groups before a fixed-shape attention kernel consumes them. Optional bounded far-history summaries can be enabled under the same interface, but the core design does not depend on them. On a 2-GPU NVIDIA A100 node, KV-RM improves mixed-length decoding throughput and tail latency relative to a static-graph baseline, reduces reserved KV memory across workload families, and removes severe burst-time latency spikes under production-trace replay. These results suggest that KV-cache movement, rather than kernel shape, can be an effective boundary for recovering runtime flexibility in static-graph LLM serving.

cs.AR↗
arXiv2026

Identification of MHD equilibrium $β$ limits for CFQS plasmas

The magnetohydrodynamic (MHD) equilibrium $β$ limits in the Chinese First QuasiAxisymmetric Stellarator (CFQS) are investigated using the NTEC code, for both the standard and the magnetic island configurations. The equilibrium $β$ limit is identified upon the onset of the rapid destruction of nested flux surfaces by evaluating several numerical metrics, including the fractal dimension, weighted Birkhoff average, and effective volume of parallel diffusion. In the standard configuration, the net-current-free and the bootstrap-current-carrying equilibria can sustain well-ordered magnetic surfaces up to $\langleβ\rangle\approx1.5\%$. The proliferation of stochastic field lines starts after the critical overlap between the internal major islands and the high-order island chains. Two types of divertor island configurations are studied based on net-current-free equilibria. It is found that the edge islands may transition into open field lines at low $\langleβ\rangle$ values and lead to a drastic shrinkage of the last closed flux surface. Meanwhile, the threshold $\langleβ\rangle$ value of the degradation of inner flux surfaces is similar to the standard configuration.

physics.plasm-ph↗