arXiv subjects

Dong Wang

Publications and source records attributed to Dong Wang.

At least 37 records · Page 2Linked to original sources

arXiv2026

Overlap-free multi-material topology optimization for minimum compliance in two and three dimensions by level-set-based negative-mapping interpolation

Gray elements and material overlaps at the interfaces limit density-based multi-material topology optimization. This article extends the level set-based negative-mapping interpolation to the multi-material proportional topology optimization of macro-scale structures in two and three dimensions. An alternating active-phase algorithm decomposes M-phase problems into two-phase sub-problems described by level set functions; an evolutionary strategy removes the need for sentitivity analysis; and the negative-mapping interpolation eliminates the overlaps of the phases in the interface elements. Cantilever and half-MBB beams in two dimensions and a cantilever beam in three dimensions, with three and four material phases, give smooth boundaries, no gray elements and overlap-free designs. The compliance lies below the SIMP value in four of the eight two-dimensional cases and 0.3%, 0.4%, 4.8% and 12.7% above it in the other four. The influence of the material properties, of the interface treatment and of the algorithmic parameters is also examined.

cs.CE↗

arXiv2026

Finer sub-Planck structures and displacement sensitivity of SU(1,1) circular states

Quantum states with sub-Planck features exhibit sensitivity to phase-space displacements beyond the standard quantum limit, making them useful for quantum metrology. In the context of the SU(1,1) group, sub-Planck features have been constructed through the superposition of four Perelomov coherent states on the hyperbolic plane (the SU(1,1) compass state). However, these structures differ in scale along different phase-space directions (anisotropic features), resulting in nonuniform sensitivity enhancement to phase-space displacements. Here, we construct $N$-component compass states, which are obtained by superposing $N \geq 6$ SU(1,1) coherent states with an even total number, evenly arranged along a circular path on the hyperbolic plane; that is, all components lie at the same distance from the origin and have equal angular spacing of $\frac{2π}{N}$. We observe that these generalized SU(1,1) compass states exhibit isotropic sub-Planck structures, leading to an isotropic enhancement in sensitivity to phase-space displacements that progressively increases with larger $N$. These states are directly relevant to quantum platforms supporting Kerr-type interactions between two bosonic modes, where the underlying SU(1,1) dynamical symmetry enables the generation of multicomponent SU(1,1) compass states. Specifically, the compact evolution of SU(1,1) coherent states under the considered dynamics enables the generation of multicomponent SU(1,1) compass states at specific times. We also investigate the effects of thermal decoherence on these multicomponent compass states within a frequency-resolved Lindblad framework, demonstrating the evolution and degradation of their nonclassical signatures.

quant-ph↗

arXiv2026

PAIR: Perceptual Affective Inference and Regulation in a Real-Time Multimodal Conversational Agent

Sustained emotional support requires generative agents to connect momentary emotion inference and regulation with continuity across encounters. We present PAIR (Perceptual Affective Inference and Regulation), a real-time multimodal agent that reconstructs how an event is appraised into an emotional state. Appraisal scaffolds produce a valence-arousal-dominance estimate and select regulation guidance, delivered through conversation with coordinated speech, color, and avatar cues. Rolling memory carries context across sessions, and the scaffold re-runs after guidance. In a 14-day deployment with 19 participants, 1,093 sessions paired initial and post-guidance estimates with unanchored self-reports. Initial valence reached MAE 1.20 on the 9-point SAM scale (r=.68), dominance reached MAE 1.30, and arousal showed weak agreement even after coarsening. Self-reported emotional change varied with initial state, with the largest valence increases in sessions that began at negative valence. Perceived understanding was associated with greater valence increase and showed little correspondence with numerical prediction error. Over two weeks, helpfulness increased while input shortened; interviews traced personalization and companionship to relevant recall, context updates, and familiar dialogue. These findings connect inference accuracy to conversational and temporal patterns of support through per-event, first-person evaluation.

cs.HC↗

arXiv2026

RoboBridge: A Self-Evolving Embodied Agent Framework for Sim-to-Real Transfer

A key challenge in bringing embodied intelligence into the real world is transferring capabilities from simulation to reality and enabling agents to continually adapt after deployment. End-to-end vision-language-action policies provide strong manipulation capabilities, but their transfer to physical environments typically relies on calibrating simulated visual and dynamical conditions, collecting additional target-domain demonstrations, and optimizing the policy through further training. Tool-using embodied agents offer flexible task orchestration, yet existing systems primarily emphasize task execution and experience reuse within a given environment, with limited support for transferring procedural knowledge and continuously adapting it across simulation and reality. We propose RoboBridge, a framework that treats sim-to-real transfer as the continued adaptation of executable task skills. The agent represents task knowledge as procedures connecting task intent, observations, tool operations, and outcome verification. Interaction feedback is used to generate candidate skill revisions, which are evaluated before being persisted or rejected. A pretrained vision-language-action policy is exposed as a reusable action tool and enhanced with inference-time guidance, enabling fine-grained execution without retraining the underlying policy. RoboBridge grounds transferable skills in task semantics and interaction interfaces shared across simulation and reality. This representation preserves reusable task structure while allowing environment-dependent operations to be selectively revised through real-world execution feedback. We evaluate the framework on LIBERO-PRO and corresponding physical tasks, studying both skill evolution and post-transfer adaptation. Our framework provides a route from one-shot policy deployment to continual procedural learning across environments.

cs.RO↗

arXiv2026

MixVLA: Adaptive Mixing of Non-Invariant Information for Generalizable Vision-Language-Action Models

Vision-Language-Action (VLA) models have achieved remarkable advances in robotic manipulation, yet their zero-shot generalization under out-of-distribution (OOD) conditions remains limited. These models often entangle task-relevant invariant structure with environment-specific non-invariant factors, causing policies to rely on spurious appearance cues during action prediction. In this work, we propose \textbf{MixVLA}, a model-agnostic training framework that improves the generalization of VLA models without requiring additional OOD data or architectural modifications. The key component of MixVLA is \textbf{Adaptive Mixing of Non-Invariant Information (AMI)}. AMI stochastically mixes non-invariant representations to regularize distribution-specific variability while preserving complementary predictive cues. The mixed non-invariant features are then fused with invariant representations for final action prediction, resulting in improved robustness without sacrificing policy expressiveness. Extensive experiments across challenging manipulation settings, including LIBERO, LIBERO-Plus, the RoboTwin perturbation suite, and real-world tasks, demonstrate that MixVLA improves overall zero-shot robustness while retaining strong in-domain performance.

cs.RO↗

arXiv2026

The Golden Path Hypothesis: Reusable Schedules in Diffusion Caching

Diffusion caching accelerates generation by replacing transformer computation with cached or predicted features at selected denoising steps. We introduce the Golden Path Hypothesis (GPH): under fixed inference conditions, prompt-independent cache schedules can achieve final-output quality comparable to the best prompt-specific schedules across prompts. We investigate the GPH across ten caching methods, four image and video models, and three cache ratios. Prompt-adaptive methods repeatedly select a small number of schedules, and reusing their most frequent schedules on new prompts closely matches the quality of prompt-specific choices. Exhaustive evaluation of 1.4 million schedules on four examples further identifies prompt-independent schedules that remain competitive on unseen prompts. To explain this transfer, we analyze denoising trajectories and the accumulation of caching errors. Latent-state trajectories exhibit similar structures across datasets and seeds, while an exact error decomposition shows that accumulated effects of earlier errors predict final latent-state error better than local approximation errors. This motivates searching for end-to-end schedules using final-output quality. With only a small set of examples, the resulting golden paths transfer across prompts and datasets, and can be tuned to the desired quality objective, including reconstruction fidelity or perceptual similarity.

cs.AI↗

arXiv2026

Auditing Agent Actions through Query-Conditioned Attribution

LLM agents increasingly take consequential actions through interactions with users, policies, and external tools. Auditing these agents requires automated attribution of realized actions to their historical basis. However, existing attribution formulations do not provide question-specific traces for diverse auditing objectives. Additionally, when access to the acting model is limited (e.g., in API-only deployments), applicable methods commonly rely on costly input perturbations or external LLM analysis of complete trajectories. We therefore formulate query-conditioned agent action attribution, a new task that takes a natural-language auditing query as input and recovers the source and ordered intermediate evidence for the query-specified aspect of an action. We instantiate this task with $A^3Bench$, a benchmark comprising 1,396 auditing queries across policy basis, parameter provenance, failure propagation, and unsafe-behavior tracing. To enable efficient, query-specific attribution, we use small open-weight models as attribution proposers that combine query-conditioned gradient saliency with query-semantic relevance to rank history units. Our proposer consistently achieves stronger source and evidence rankings at lower inference cost than open-weight baselines, improving source MRR by up to 40.9\% and evidence MAP by 42.1\% with only two forward passes and one backward pass. Controlled evaluations confirm that our proposer improves attribution specificity by adapting its rankings to fine-grained changes in the auditing query. Building on a proposer ensemble, our end-to-end system surpasses the strongest frontier-model baseline in source accuracy (64.5\% vs.\ 60.4\%) while reducing empirical deployment latency by 29.9\% relative to the fastest frontier API baseline. Code and data will be released after the initial review period following final validation and cleanup.

cs.AI↗

arXiv2026

SIPO: Unifying Reinforcement Learning with On-Policy Self-Distillation

Reinforcement learning with verifiable rewards (RLVR) has become a standard paradigm for improving large language models (LLMs) on various tasks, yet its sparse outcome rewards lack token-level credit assignment for intermediate steps. To address this, on-policy self-distillation (OPSD) leverages a self-teacher with privileged context to provide additional dense learning signals. However, because the self-teacher is often overconfident and imposes excessive penalties on long reasoning trajectories, OPSD frequently struggles in practice. To mitigate this, we propose self-instructing policy optimization (SIPO) with a contrastive self-teacher to provide dense credit. At each iteration, SIPO samples multiple rollouts per prompt from the current policy, scores them with environment rewards, and constructs two teacher contexts for each rollout by pairing the reference answer with mistakes made within the group. The model then re-evaluates its own responses under both contexts, using the difference between the two teacher log-probabilities as token-level feedback, so that biases shared by both contexts are expected to largely cancel. The resulting objective yields a token-level advantage for every rollout: the reward still sets the main direction of each update while the self-teacher redistributes credit across tokens. Even in groups where every rollout fails and group-relative advantages vanish, SIPO still provides a learning signal. By preserving direct optimization of the task reward while providing dense, token-level feedback, this approach bridges reinforcement learning and on-policy self-distillation. Extensive experiments across multiple reasoning and code-generation benchmarks demonstrate that SIPO outperforms both RLVR and OPSD baselines without an external teacher or additional generation.

cs.AI↗

arXiv2026

When Upstream Messages Override Correct Answers: A Controlled Study of Multi-Agent LLM Collaboration

Multi-agent LLM systems rely on message passing among specialized agents to accomplish complex tasks. However, an upstream agent may provide useful information or an incorrect answer that causes a downstream agent to override a correct answer supported by its own evidence. Prior work has not clearly separated the benefits of communication from the damage caused by incorrect messages. We study this problem with controlled experiments across five benchmarks and five receivers, keeping the downstream task and evidence fixed while comparing answers under three conditions: no message, the upstream agent's original message, or a message with the opposite conclusion. Our experiments reveal three key findings. First, messages often help when the downstream agent would otherwise answer incorrectly. Second, messages can also hurt: when the downstream agent would answer correctly without a message, an incorrect upstream message changes the answer in up to 32% of cases. Third, in 94% of audited harmful cases, the downstream agent copies the upstream's specific wrong answer--a pattern we term answer substitution. Removing unreliable messages recovers part of the lost accuracy, suggesting that communication should be selective based on upstream reliability and the evidence already available to the downstream agent.

cs.AI↗

arXiv2026

DiffReID: Discriminative Diffusion Model for Object Re-Identification

As a fundamental image processing task, object Re-Identification (ReID) aims to retrieve objects across non-overlapping cameras. Recently, with the development of deep learning, significant advancements have been made in object ReID. However, most existing methods suffer from generalization due to the limited size and diversity of ReID datasets. Meanwhile, current models tend to focus on extracting semantic patterns rather than learning identity-aware feature distributions. To address these issues, we propose a novel feature learning framework named \textbf{DiffReID} for object ReID. It leverages a discriminative diffusion model to gradually learn identity-aware distributions and generate identity-invariant features. More specifically, with the Contrastive Language-Image Pre-training (CLIP) model, we first obtain identity-aware text features by prompt tuning. Then, we propose a Vision-guided Noise Generator (VNG) to initialize probabilistic noises and gradually corrupt identity-aware text features. Afterwards, we take visual features as conditions and propose a Light Weight Denoiser (LWD) to denoise the corrupted text features step-by-step for identity-aware distribution learning. To obtain discriminative features, we further generate identity-invariant guided features from randomly sampling visual-guided noises. Finally, we propose a Mutual Enhancement Constraint (MEC) to facilitate mutual learning between visual features and guided features to enhance the representation robustness and discrimination. Extensive experiments on five object ReID benchmarks demonstrate that our method shows better results than most state-of-the-art methods. The source code is available at https://github.com/AWangYQ/DiffReID.

cs.CV↗

arXiv2026

Nonlinear stability of 3-D plane Couette flow for Boussinesq system in a finite channel

In this paper, we study the hydrodynamics stability of the 3-D plane Couette flow $v_s=(y,0,0)$ with constant temperature state $ρ_s$ for the Boussinesq system in a finite channel $\mathbb{T}\times [-1,1]\times \mathbb{T}$ with non-slip boundary condition in $y$ direction. We prove that if the initial velocity field $v_{in}$ and initial temperature $ρ_{in}$ satisfy \begin{align*} \|v_{in}-(y,0,0)\|_{H^2} \leq c_0 ν, \ \ \|ρ_{in}-ρ_s\|_{H^2} \leq c_0 ν^2, \end{align*} for some $c_0>0$ independent of the viscosity and thermal diffusion $ν$, then the solutions to 3D Boussinesq system do not transit from the Couette flow and constant temperature state. To our best knowledge, this is the first result about the nonlinear stability of plane Couette flow for 3-D Boussinesq system in the domain with non-slip boundary condition. This result, together with [18], implies that the strong boundary layer effect does not lead to strong instability.

math.AP↗

arXiv2026

Beyond Verbalized Confidence: Calibrating Reasoners with Differentiable Readouts

Reinforcement learning with verifiable rewards (RLVR) trains reasoning models to produce correct answers, but does not ensure that their stated confidence is calibrated. The resulting models are systematically overconfident. Recent methods train calibration inside the RLVR loop by having the model state a numerical confidence alongside its answer, but they all obtain the confidence by sampling it as text. This choice imposes two costs: a sampled confidence introduces variance and in practice collapses to a handful of distinct values, and sampling makes the confidence non-differentiable, forcing the calibration loss through a scalar reward. We propose CREDO (Confidence REaDOut) to replace sampling with a deterministic readout. While RLVR optimizes correctness, CREDO reads the confidence from a dedicated token pair in the model's output distribution and trains it by differentiable regression. CREDO further turns the trained confidence into a signal for accuracy, weighting rollouts by how far confidence and outcome disagree, so that accuracy and calibration improve together. Across mathematical and code reasoning, CREDO attains the best accuracy and calibration, and the gains extend to abstention and selective prediction.

cs.LG↗

arXiv2026

Learning to Track from Privileged Target Appearances

Target templates define what a visual tracker searches for, yet the templates available at inference trade off localization certainty with appearance freshness: the initial ground-truth template is exact but becomes stale, whereas recent templates better reflect the current appearance but are cropped from uncertain predictions. We quantify this bottleneck with a non-deployable oracle that supplies an exact current-frame target crop, improving AUC on LaSOT by 15.2 percentage points. This gap reveals a training-only opportunity: frame-level ground truths provide exact current- and future-frame target crops, although such crops are unavailable at deployment. We introduce Privileged Appearance Transfer for Tracking (PATT), a teacher-student training framework that transfers these privileged appearances to a deployable tracker through multi-level representation prediction. The privileged teacher observes exact target crops from past, current, and future frames, whereas the student receives only past-frame templates and learns to predict the teacher's search representations. To avoid transferring unreliable teacher signals, PATT weights this transfer by the teacher's relative localization advantage over the student and its absolute localization accuracy. After training, the teacher, latent predictor, reliability weights, and privileged crops are removed, leaving standard student-only inference. Across seven benchmarks at two model scales, PATT achieves consistent gains under both long- and short-term tracking protocols.

cs.CV↗

arXiv2026

UCON: Uncertainty-aware Navigation with Historical Re-association in Dynamic Environments

Autonomous navigation in dynamic environments is hindered by two fundamental challenges: perception instability and uncertainty-optimization mismatch. The former leads to identity switches and unreliable motion estimation, while the latter prevents principled incorporation of motion uncertainty into trajectory optimization. To address these challenges, we propose UCON, an uncertainty-aware navigation algorithm in dynamic environments. For perception instability, we present a point-level historical re-association mechanism that leverages historical point cloud fragments to recover lost targets while maintaining identity continuity. Subsequently, a Kalman filter is employed to provide anisotropic motion state estimation and covariance propagation. To resolve the uncertainty-optimization mismatch, we transform predicted states and their covariances into uncertainty sectors, which are embedded as differentiable cost terms within a trajectory optimization framework. This achieves consistent uncertainty-aware dynamic obstacle avoidance while maintaining smoothness and feasibility. Extensive simulations and real-world experiments demonstrate that, while maintaining high computational efficiency, UCON achieves superior perception stability and robust navigation performance in dynamic environments compared to state-of-the-art methods. The code will be open-sourced to facilitate further research.

cs.RO↗

arXiv2026

Scheduling Coflows in Multi-Core OCS Networks with Performance Guarantee

The coflow abstraction captures application-level communication patterns and enables coordinated scheduling of parallel flows to reduce job completion times in distributed systems. Modern data center networks (DCNs) are employing multiple independent optical circuit switching (OCS) cores operating concurrently to meet the massive bandwidth demands of application jobs. However, existing coflow scheduling research primarily focuses on the single-core setting, while studies of multi-core fabrics have largely considered electrical packet switching (EPS) networks. To address this gap, this paper studies the coflow scheduling problem in multi-core OCS networks under the not-all-stop reconfiguration model, in which the reconfiguration of one circuit does not interrupt other circuits. The challenges stem from two aspects: (i) cross-core coupling induced by traffic assignment across heterogeneous cores; and (ii) per-core OCS scheduling constraints, namely \textit{port exclusivity} and \textit{reconfiguration delay}. We propose an approximation algorithm that jointly integrates cross-core flow assignment and per-core circuit scheduling to minimize the total weighted coflow completion time (CCT) and establish a provable worst-case performance guarantee. Furthermore, our algorithm framework can be applied to the multi-core EPS scenario with a corresponding approximation guarantee for packet-switched fabrics. Trace-driven simulations using real Facebook workloads demonstrate that our algorithm can reduce the total weighted CCT and tail CCT.

cs.DC↗

arXiv2026

Hunyuan-A13B Technical Report

We present Hunyuan-A13B, an open-source large language model based on a Mixture-of-Experts architecture. It contains 80 billion total parameters but activates only 13 billion during inference, balancing model capability, computational efficiency, and deployment cost. The model is pretrained on a rigorously filtered 20T-token corpus with enhanced STEM data curation, improving factual reliability and reasoning ability. High-quality supervised fine-tuning and large-scale reinforcement learning further enhance its overall performance. Hunyuan-A13B also introduces a dual-mode Chain-of-Thought framework that adapts reasoning depth to task complexity: fast thinking for routine queries and slow thinking for complex, multi-step problems. Evaluations show competitive performance across mathematics, science, programming, general language understanding, and agent tasks, often approaching that of much larger models. Its high inference throughput makes it suitable for latency-sensitive applications. We release Hunyuan-A13B to support open research and practical LLM deployment.

cs.AI↗

arXiv2026

Hard to soft edge transition for the Muttalib-Borodin ensembles with integer parameter $θ$

We find the universal limiting correlation kernels of the Muttalib-Borodin (MB) ensembles with integer parameter $θ\geq 2$ at $0$ in the transitive regime between the hard edge regime and the soft edge regime. This generalizes the previously studied hard edge to soft edge transition in unitarily invariant random matrix theory by Its, Kuijlaars and Östensson, which is the $θ= 1$ special case of our MB ensemble. The derivation is based on the vector Riemann-Hilbert (RH) problems for the biorthogonal polynomials associated with the MB ensemble. In the analysis of the RH problems, we construct matrix-valued model RH problems of size $(θ+ 1) \times (θ+ 1)$, and prove the solvability of the model RH problems by a vanishing lemma. The new limiting correlation kernels are proved to be universal for a large class of potential functions, and they interpolate the Meijer G-kernels for the hard edge regime and the Airy kernel for the soft edge regime. We observe that the new limiting correlation kernels have the integrability that is not seen in previous studies in random matrix theory and determinantal point processes. In the $θ= 2$ case, we give a detailed analysis of the Lax pair associated with the model RH problem, show that it results in the Chazy I equation, which has a Painlevé IV reduction, and find that the Lax pair is in the Drinfeld-Sokolov hierarchies.

math-ph↗

arXiv2026

WOLF: World Model Guided LiDAR Exploration with Predictive Frontiers

LiDAR-based unmanned aerial vehicle (UAV) exploration builds maps by continually selecting where to observe next. However, decisions based on the measured map provide limited foresight into spatial continuations behind occlusions, leaving potentially informative directions unrecognized. We present WOLF, a world-model-guided framework that predicts future observations to enhance autonomous exploration. In the training stage, a recurrent world model learns observation dynamics from exploration trajectories, with recurrent memory retaining the spatial context needed to interpret partial observations across successive views. Building on this context, the model combines observation history with candidate motions during exploration to predict local occupancy and visibility. To guide further sensing, a predictive frontier generation mechanism then aligns and fuses these predictions using confidence, branch agreement, and observation quality to identify promising regions. The resulting predictive frontiers join measured ones to guide geometric viewpoint selection and trajectory generation, while new scans update subsequent predictions. In simulations, our method reduces mean terminal time by 10.9% relative to EPIC in Garage at comparable coverage and increases mean coverage from 42.12% to 98.35% in Tunnel. Real-world experiments further demonstrate onboard deployment of the learned model for online inference during physical flight.

cs.RO↗