arXiv subjects

Yu Zhang

Publications and source records attributed to Yu Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Behavioral Probes for Information Flow in LLM Swarms

Most existing approaches to LLM swarm optimization are utility-driven, focusing on final task utility while leaving the processing, integration, and reliance of information within individual interactions largely implicit. To address this, we study LLM swarm interactions from an information-flow-based view and introduce behavioral probes to quantify information flow patterns. Specifically, we repurpose External Context Score (ECS) and Parametric Knowledge Score (PKS)---originally designed for hallucination analysis---to measure a node's external-context dependence and parametric-knowledge dependence, respectively. Across controlled and realistic swarm settings, we observe consistent patterns: ECS increases with input relevance and distinguishes useful multi-source contributions, while PKS decreases as informative context accumulates. Building on this heuristic interpretation, we demonstrate how these probes can complement utility-driven structure optimization in two controlled applications. Probe-guided node pruning reduces redundant communication and token consumption, and probe-guided reasoning-path search accelerates convergence compared to utility-only baselines. Our empirical results suggest that behavioral probes can make the internal information flow of LLM swarms observable and serve as effective auxiliary signals for structure optimization.

cs.MA↗

arXiv2026

CARNet: End-to-End Coupled Association-Recognition Network for Fine-Grained Multi-Drone Collaborative Detection

Multi-drone collaborative detection aims to enhance target recognition by leveraging complementary observations from overlapping aerial viewpoints. Existing approaches typically follow a two-stage pipeline that first establishes cross-view correspondences and then performs fusion-based recognition. However, this decomposition suffers from inconsistent optimization objectives, redundant pairwise modeling, and irreversible error propagation, where incorrect associations inevitably undermine subsequent recognition performance. To overcome these limitations, we propose the Coupled Association-Recognition Network (CARNet), a unified framework that jointly optimizes cross-view target association and recognition fusion through a shared target-pair representation. To improve association reliability in densely populated scenes with visually similar targets, CARNet explicitly incorporates spatial cues into pairwise relation modeling. Moreover, we introduce two complementary causal-inspired intervention strategies: the first enhances cross-view association by emphasizing appearance-confusing pairs, while the second bolsters the robustness of fusion-based recognition by learning stable multi-view representations under incomplete visual evidence. In addition, we present MDMT-CD, a fine-grained extension of the MDMT dataset that enriches the original vehicle annotations with detailed category labels, providing the first benchmark specifically designed for multi-drone collaborative vehicle detection. Extensive experiments demonstrate that CARNet consistently outperforms existing state-of-the-art methods. The MDMT-CD dataset and source code will be available at https://github.com/bigdreaming11/CARNet.

cs.CV↗

arXiv2026

MMedAgent-Plus: Scaling Multimodal Medical Agents across Modalities, Tasks and Workflows

Medical workflows often require multiple interdependent analyses, but existing specialist and generalist AI models are not designed to coordinate them. We introduce MMedAgent-Plus, a multimodal medical tool-use agent that combines a vision-language planner with 18 leading specialist models spanning 12 tasks and 8 modalities. Through a unified tool interface, the planner selects specialists, executes them in sequence, and synthesizes responses grounded in their outputs. The planner is instruction-tuned on 208,928 samples from 32 public medical benchmarks, covering single- and multi-tool interactions across single and multiple rounds. We separately evaluate tool orchestration, response synthesis, and end-to-end execution on held-out samples. MMedAgent-Plus achieves tool-call accuracies of 91.0%, 98.1%, and 80.7% in single-tool multi-round, multi-tool multi-round, and multi-tool single-round settings, respectively. Given correct tool outputs, MMedAgent-Plus achieves scores of 14.4-14.8 out of 15 for faithfulness, completeness, and conciseness across 600 evaluation samples. In end-to-end execution without pre-supplied tool outputs, MMedAgent-Plus correctly invokes the required specialist in 90.6% of 350 evaluation cases spanning seven tasks. Its final-answer accuracy reaches 0.98, matching the dedicated leading specialist and exceeding the strongest baseline (0.28).

cs.CL↗

arXiv2026

PhysLDM: Latent Diffusion for High-Fidelity Deformable Simulation

Neural simulation of high-fidelity deformable bodies is a foundational challenge in computer graphics and physical AI. Long-horizon prediction for high-resolution 3D volumetric meshes is hard: autoregressive methods are prone to error accumulation, while direct multi-frame prediction at native resolution is computationally prohibitive. This motivates a compact spatiotemporal latent representation, which is largely unexplored for mesh-based volumetric physics. Meanwhile, it remains unclear whether deterministic regression or generative diffusion is the more suitable predictive paradigm. To address these coupled challenges, we introduce PhysLDM, a unified latent-diffusion paradigm for one-shot deformable simulation. Its core is a holistic spatiotemporal VAE that avoids the "staircase" artifacts of standard temporal compression (e.g., video VAEs), achieving ~2.48 mm reconstruction precision on meter-scale scenes at up to 78x token compression. Based on this reliable latent space, we systematically compare regression and diffusion methods. Our experiments uncover a key modeling insight: complex deformable dynamics are often chaotic, and in this regime deterministic regression tends to produce non-physical averages, whereas diffusion better models their distribution. Accordingly, we employ a latent diffusion model that effectively learns from chaotic data to generate physically plausible trajectories. Trained purely kinematically on an Objaverse-scale dataset, a single PhysLDM generalizes zero-shot to unseen OOD datasets. Its differentiability further enables efficient solution of inverse problems and higher-order design optimization. To our knowledge, PhysLDM is the first high-fidelity spatiotemporal autoencoder and latent-diffusion paradigm for volumetric deformable dynamics, offering a scalable and robust approach to neural simulation. Project Page: https://yucrazing.github.io/physldm/

cs.CV↗

arXiv2026

Co-adaptive Rule Routing and Constraining for Structured Knowledge Retrieval

Structured knowledge retrieval acquires task-relevant evidence from structured knowledge sources for downstream tasks. Existing methods rely on unconstrained retrieval or predefined query-rule supervision, such as author-paper rules for scientific discovery and anatomy-drug rules for precision medicine. Such supervision is rarely available in practice. We propose Co-Adaptive Rule Routing and Constraining Retrieval (CoRE), which jointly learns which reasoning rule to assign to each query and how to retrieve evidence under that rule, without predefined query-rule pairs. A routing agent selects a rule, and its corresponding retriever traverses the knowledge base and ranks evidence that satisfies the rule's constraints. Because routing and retrieval are interdependent, CoRE alternates between reinforcement learning for the routing policy and contrastive learning for the rule-constrained retriever. The selected rules provide supervision for retriever training, while retrieval performance rewards the routing policy. This closed-loop process enables both components to co-adapt without explicit query-rule supervision. Experiments on STaRK-QA datasets across scientific discovery, biomedical reasoning, and e-commerce knowledge management demonstrate CoRE's effectiveness. Ablations show the benefit of co-adapting the router and retriever. Generation-based routing transfers across datasets and benefits from multi-domain pretraining followed by target-specific fine-tuning. Analysis of rule usage across queries also informs automatic rule refinement and discovery.

cs.IR↗

arXiv2026

Sink-Sparse Hybrid Attention for Efficient Inference in T2I/T2V Diffusion Transformers

Diffusion Transformers (DiTs) have become a common backbone for high-quality text-to-image and text-to-video generation, but repeated dense self-attention remains a major inference bottleneck as token sequences grow and the same operator is invoked across many denoising steps. Our study suggests that dense attention in pretrained DiTs contains reusable layer/head-specific high-mass structure that can be identified from a small offline calibration set and reused across future generations. This leads to Sink-Sparse Hybrid Attention, a train-free attention replacement that reuses static sink masks and supplements them with a lightweight pooled-KV branch, while leaving model weights and sampling schedules unchanged. On Wan2.1-T2V-1.3B, the method attains 1.54$\times$ end-to-end speedup while maintaining generation quality. On Z-Image-Turbo, it attains 1.57$\times$ end-to-end speedup with competitive GenEval2 performance. These results indicate that reusable importance-aware sparse support can provide an effective train-free acceleration path for both T2V and T2I DiTs.

cs.CV↗

arXiv2026

VulnSAGE: Adjudicating SAST Reports by Connecting LLM Reasoning to Program Structure

Static Application Security Testing (SAST) reports flag operations whose safety depends on project-specific context, and maintainers must decide which reports describe real defects. Large language models (LLMs) can interpret this context, but adjudicating a report with them requires the decisive code, which often lies far from the reported line, and a decision basis that program analysis can check. We present VulnSAGE, an adjudication system for SAST reports that connects LLM reasoning to program structure in both directions. Its report-centered slicer recovers the dependencies of the reported operation together with the program structure that explains them, such as the caller object on which a helper acts. CLAIM, an intermediate representation for report judgments, records each model interpretation as a semantic fact that cites the program relations it assumes. Program analysis checks these premises, and the verdict follows from the facts whose premises hold. Both mechanisms share a program graph, which lets the model cite the code it reads and lets the checker query those citations. Statically supported true positives then guide proof-of-concept generation and runtime reproduction. Across 187 labeled reports from four corpora covering five vulnerability families, VulnSAGE rejects 89 of 97 false alarms and retains 85 of 90 real defects. Under a common PoC procedure, 25 of its positive verdicts reproduce at runtime, compared with three for the LLM-based adjudicator LLM4SA. Applied to 45,440 alerts raised by five SAST tools on the 200 most-starred C/C++ projects on GitHub, it reproduces 104 defects at runtime. Maintainers have confirmed 24 of the 26 defects they have responded to, including ten previously unknown defects, nine of which are already fixed.

cs.SE↗

arXiv2026

Knowledge Graphs as Files: An Agent Harness Where Tool Creation and Tool Use Co-Evolve

Searching text-rich knowledge graphs requires following relations while examining textual evidence. Predefined graph interfaces constrain what evidence agents can access, even when they write code to combine tool calls or process retrieved subgraphs. We propose direct knowledge graph interaction (DKGI), a harness in which language model agents create graph access tools directly over complete node and edge files. The agent writes Python code to follow relations, combine node sets, and select textual evidence, interleaving execution with shell search and file inspection. This lets agents process large node sets in bulk while returning selected text for the model to assess. Execution feedback guides tool creation and revision, while the evolving tools determine what evidence the agent gathers next. On benchmark datasets spanning three domains, training-free DKGI exceeds the strongest published baseline by at least 6.7 Hit@1 points on every graph, even with a minimal harness. Controlled ablations show the benefits of multi-turn programming over shell-only access and predefined graph tools, while trajectories show agents revising code to extend searches or address errors. These findings highlight the value of allowing tool creation and tool use to co-evolve during graph search. Our code and data are available at https://github.com/leopoldwhite/DKGI.

cs.CL↗

arXiv2026

CA-OPD: Confidence-Aware On-Policy Distillation for Structured Visual Prediction

Autoregressive vision language models unify heterogeneous perception tasks but are highly susceptible to compounding errors. On-policy distillation (OPD) bridges the training-inference mismatch by training students on their own rollouts. However, unreliable student predictions, especially early in training, can derail the trajectory and degrade the quality of teacher supervision. While recent interleaved distillation methods allow the teacher to verify and replace student tokens, they primarily rely on rigid ranking metrics rather than exact teacher confidence, and they overlook how intervention decisions can inform token-level supervision. To address this, we introduce Confidence-Aware On-Policy Distillation (CA-OPD), a framework that couples reliable rollout construction with adaptive supervision. CA-OPD utilizes teacher confidence to selectively correct unreliable student transitions, gradually transferring rollout control to the student via a strict-to-relaxed schedule. Crucially, CA-OPD aligns knowledge transfer with these intervention decisions: corrected positions receive direct cross-entropy supervision from the teacher's prediction, while retained positions benefit from the teacher's full predictive distribution. Evaluated in a multi-teacher setting for GUI grounding and optical character recognition, CA-OPD substantially improves the Qwen3.5-0.8B baseline across all six target benchmarks, including gains of $9.50$ points on ScreenSpot-Pro and $6.72$ points on OCRBench-v2 English. Controlled studies further show that the gains depend on intervention placement, progressive rollout control, and intervention-aligned supervision, rather than intervention frequency alone.

cs.CV↗

arXiv2026

VGGT-Diff: Visual Geometry Meets Diffusion for Sparse-View Novel View Synthesis

We present VGGT-Diff, a geometry-routed multi-view diffusion model for sparse-view novel view synthesis. Existing novel view synthesis (NVS) methods face a fundamental trade-off: reconstruction-based approaches preserve observed geometry but struggle to synthesize unseen regions, while diffusion-based methods provide strong generative priors yet rely on implicit source-to-query correspondence. VGGT-Diff bridges these regimes by routing visual geometry latents from VGGT-Ω into a pretrained video diffusion model. Each visual token is associated with a 3D point and confidence, then transformed into query-aligned latent conditions through a confidence-aware Visual Geometry Router (VGR) that preserves front and back surface evidence. These conditions guide joint target-view denoising, while Point-Track Residual Consistency (PTRC) regularizes predicted-clean residuals along reliable 3D tracks, improving multi-view stability. We further introduce robust geometry conditioning, combining training-time regularization with inference-time guidance for improved robustness. Experiments show competitive or state-of-the-art performance across interpolation and extrapolation under different viewpoint difficulties. Our code is available at https://github.com/chenkangjie1123/VGGT-Diff.

cs.CV↗

arXiv2026

Selective Successor Eligibility Traces for RLVR

Reinforcement Learning with Verifiable Rewards (RLVR) has become a key approach for improving the reasoning abilities of large language models. However, widely used critic-free algorithms such as Group Relative Policy Optimization (GRPO) suffer from coarse-grained temporal credit assignment, as they assign the same trajectory-level advantage to all tokens and fail to distinguish the contributions of different reasoning steps. To address this challenge, we propose Selective Successor Eligibility Traces (S-trace), an critic-free eligibility trace method that enables more effective credit assignment. S-trace introduces successor importance reweighting to recover downstream credit signals under PPO-style clipping and propagates them to selected historical tokens through entropy-based token selection. Theoretically, we establish an upper bound on its discrepancy from the raw eligibility traces-based policy gradient estimator, jointly controlled by the KL divergence and the trace decay factor $λ$. Experiments show that S-trace achieves absolute gains in average pass@16 of 1.80\% and 6.17\% over GRPO, and 2.32\% and 0.87\% over trace-style GRPO($λ$), on Qwen3-1.7B and Qwen3-4B, respectively, while maintaining robust gains on Qwen3-8B and producing shorter responses.

cs.LG↗

arXiv2026

REINS: Agent-guided Thread Scheduling for Concurrency Bug Detection and Reproduction

Concurrency bugs manifest only under specific thread interleavings, making them hard to detect and reproduce. Existing controlled concurrency testing techniques take over thread scheduling to explore the interleaving space. As the space grows rapidly with program scale, finding interleavings that can trigger bugs within a limited budget remains challenging. The program's concurrency semantics can reveal which variable accesses and synchronization operations under specific interleavings lead to unexpected program behavior, and thus help focus the exploration on bug-relevant interleavings. However, prior techniques have not fully exploited such semantic information. In this paper, we present REINS, an agentic approach for detecting and reproducing concurrency bugs. At its core, REINS guides thread scheduling with the agent's reasoning about the concurrency semantics and known concurrency bug patterns. REINS starts with runtime observations and combines them with static information to reason about the program's concurrency behavior and formulate hypotheses about potential bugs. A hypothesis translates into a set of constraints on event orders and variable states, guiding a controlled scheduler toward interleavings likely to trigger the bug. REINS draws conclusions from the execution results and keeps refining the constraints until the bug is triggered or the hypothesis is refuted. Compared with existing techniques on three public benchmarks, REINS detects more bugs within the same time budget. Furthermore, REINS found six previously unknown concurrency bugs in real-world projects, three of which have been confirmed by the developers.

cs.SE↗

arXiv2026

Direct view of exciton and carrier dynamics in twisted bilayer graphene

Twisted bilayer graphene (tBLG) on H-intercalated SiC (twist angle of $9 \pm 1^{\circ}$) is studied by position\mbox{-}, time\mbox{-}, and angle-resolved photoemission spectroscopy (ARPES). Both graphene layers are hole doped, with a Dirac-point offset of 30~meV. Time-resolved ARPES near the van Hove singularity reveals a spectral feature within the hybridization-induced band gap, together with excited electrons in the conduction band. The in-gap feature exhibits distinct dynamics, and its binding energy and $k$-space location are consistent with the proposed non-hydrogenic ``ghost-Fano'' exciton arising from destructive interference between two nearly degenerate excitonic channels. Near the Dirac points, the twist separates the largely layer-polarized Dirac cones in momentum space, providing direct layer selectivity. We find markedly different peak electronic temperatures in the two graphene layers, even though their subsequent cooling dynamics are very similar. We attribute this to an ultrafast interlayer redistribution of electronic energy, driven by an interlayer electric field that arises from the different doping in the layers.

cond-mat.str-el↗

arXiv2026

Born-Oppenheimer chiral effective field theory and its application to $X_b$

Following the discovery of the exotic state $X(3872)$ in the charmonium spectrum, the possible existence of an analogous state, $X_b$, in the bottomonium sector has attracted considerable attention of the hadronic physics community. In this work, we lay the ground for the development of Born-Oppenheimer chiral effective field theory (BO$χ$EFT), that allows us to include both quark and hadron degrees of freedom as well as the full complexity of the meson-exchange potential in a systematic way. For illustration, the formalism to leading order (LO) is applied to the system of $χ_{b1}$ states and in particular the $X_b$. The two parameters that are present at this order are constrained using experimental information on the bottomonium spectrum together with lattice input on static quark-antiquark potentials, including string breaking effects. The framework identifies the observed $1^{++}$ state at 10.513 GeV with a predominantly molecular $X_b$, the bottom counterpart of the $X(3872)$, and predicts an additional predominantly compact bottomonium state at 10.544 GeV, about 60 MeV below the $B\bar{B}^*$ threshold.

hep-ph↗

arXiv2026

FrameScout: Scouting Query-Relevant Frames for Long Video Understanding

Long video understanding requires VideoLLMs to answer queries about videos spanning tens of minutes to hours, but such long videos produce far more visual tokens than current VideoLLMs can process. To address this issue, keyframe selectors are proposed to select the most query-relevant frames by computing the similarity between frame and query embeddings. Existing selectors either encode frames independently without temporal context or process chunks in isolation with ranking objectives that are difficult to optimize. To alleviate those limitations, we propose FrameScout with two core designs: (i) a streaming selector architecture that leverages a sliding-window KV cache and successor aggregation to produce temporally-contextualized frame embeddings; (ii) a frame-query contrastive objective that aligns frame and query embeddings in a shared space, directly training the selector to distinguish relevant frames from irrelevant ones. Extensive experiments on three long-video benchmarks and six base VideoLLMs demonstrate that FrameScout consistently achieves state-of-the-art performance. Code is available at https://github.com/Hubert-1102/FrameScout.

cs.CV↗

arXiv2026

A Concept-informed Interpretable Artificial Intelligence Framework for Breast Ultrasound Diagnosis

Most existing artificial intelligence systems for breast ultrasound rely primarily on image-driven prediction, with limited incorporation of the knowledge from the Breast Imaging Reporting and Data System (BI-RADS) and the intermediate assessment processes underlying radiologists' workflow, thereby constraining their interpretability and alignment with clinical practice. We curated, to our knowledge, the largest multicenter collection of breast ultrasound images paired with structured BI-RADS descriptors, comprising 9,941 patients across 20 independent datasets. Across 14 testing datasets, IBUS demonstrated consistent generalization performance across different clinical centers and imaging conditions. Incorporating BI-RADS descriptors enabled fine-grained risk stratification, while cross-checking predicted BI-RADS grades against malignancy probabilities identified internally discordant predictions that may warrant additional clinical review. In a reader study involving eight radiologists, IBUS assistance increased diagnostic sensitivity among both junior and senior readers, with the potential of guiding radiologists to make targeted revision in risk stratification. Overall, IBUS exhibits its potential as a second-stage decision-support system that complements radiologists by providing additional risk estimates and interpretable evidence for diagnostically challenging cases. The demo video of IBUS platform is publicly available at https://github.com/zsz1997/IBUS.

cs.CV↗

arXiv2026

EmbodiedHead: Real-Time Listening and Speaking Avatar for Conversational Agents

We present EmbodiedHead, a speech-driven talking head framework that equips LLMs with real-time visual avatars for conversation. A practical embodied avatar must achieve real-time generation, unified listening-speaking behavior, and high rendered visual quality simultaneously. Our framework couples the first Rectified-Flow Diffusion Transformer (DiT) for this task with a differentiable renderer, enabling diverse, high-fidelity generation in as few as four sampling steps. Prior listening-speaking methods rely on dual-stream audio, introducing an interlocutor look-ahead dependency incompatible with causal user--LLM interaction. We instead adopt a single-stream interface with explicit per-frame listening-speaking state conditioning and a Streaming Audio Scheduler, suppressing spurious mouth motion during listening while enabling seamless turn-taking. A two-stage training scheme of coefficient-space pretraining and joint image-domain refinement further closes the gap between motion-level supervision and rendered quality. Extensive experiments demonstrate state-of-the-art visual quality and motion fidelity in both speaking and listening scenarios. Project page: https://vpx-ecnu.github.io/EmbodiedHead-Website/.

cs.CV↗

arXiv2026

AdaGamma: State-Dependent Discounting for Temporal Adaptation in Reinforcement Learning

The discount factor in reinforcement learning controls both the effective planning horizon and the strength of bootstrapping, yet most deep RL methods use a single fixed value across all states. State-dependent discounting is a natural way to adapt temporal credit assignment to local state structure, but naively learning a discount function inside deep actor--critic methods creates a self-referential target: the learned discount scales the same bootstrap term against which it is trained. Consequently, TD-error-based objectives can admit a degenerate shortcut, reducing target dependence on downstream value estimates instead of learning meaningful temporal structure. We propose \textbf{AdaGamma}, a practical framework for state-dependent discounting in deep actor--critic reinforcement learning. AdaGamma learns a bounded discount function $γ_ϕ(s)$ and trains it using a return-consistency objective that matches a one-step adaptive bootstrap to a multi-step return computed with a lagged, stop-gradient reference discount. This removes the direct target manipulation path and encourages the discount to encode state-dependent value propagation. We analyze the Bellman operator induced by a fixed state-dependent discount in finite tabular MDPs, proving contraction, exact soft policy iteration, and Lipschitz sensitivity under standard boundedness assumptions. Empirically, AdaGamma integrates into SAC and PPO, avoids TD-collapse, improves continuous-control benchmarks, and achieves statistically significant gains in a four-week online A/B test on the JD Logistics platform.

cs.LG↗