Search arXiv⌕ Search

arXiv subjects

Tao Zhang

Publications and source records attributed to Tao Zhang.

At least 19 recordsLinked to original sources
arXiv2026

Active tissue mechanics drives a diversity of collective flows in uniaxially compressed 3D spheroids

Compression can enhance or suppress cell motion and induce coherent rotation in tissue spheroids, but the physical factors governing these responses remain unclear. Using a minimal three-dimensional active vertex model, we examine how tissue mechanics, active driving, and directional alignment shape the response to uniaxial compression. Without alignment, compression enhances transverse motion in solid-like spheroids throughout the tested activity range, whereas fluid-like spheroids switch from acceleration at low activity to slowing at high activity. Changes in contact persistence connect these responses to neighbor exchange. Passive five-cell calculations show that compression reshapes local rearrangement barriers depending on geometry and loading direction; a high-strain increase in the fluid-like mean coexists with many low-barrier paths. With strong alignment, compression induces coherent rotation in solid-like spheroids, while fluid-like spheroids show no comparable rotation. Rotating solids retain more persistent cell contacts than fluids and develop tangential organization of peripheral cells. Controlled perturbations reveal that compression softens a differential-twist mode and activity slows its relaxation. Together, these results link activity-dependent reversal of the speed response and compression-induced coherent rotation to contact dynamics, local rearrangement costs, and active driving. Our findings demonstrate that one minimal model, with parameters representing effective mechanical and active tissue properties, captures diverse collective responses to compression and provides a framework for predicting how different tissues respond to mechanical loading.

cond-mat.soft↗
arXiv2026

Spatial-Frequency Learning for Scanning Transmission Electron Microscopy Image Enhancement

Scanning Transmission Electron Microscopy (STEM) enables the observation of atomic arrangements at sub-angstrom resolution and plays a key role in the discovery and characterization of advanced materials. Raw STEM images are often noisy and blurred due to multiple degradations. Although recent deep learning methods improve STEM images, their performance remains limited under severe degradations. In this work, we propose a spatial-frequency learning framework for STEM image enhancement. Specifically, we design a spatial-frequency fusion network that exploits periodic atomic arrangements in the frequency domain and restores local atomic details in the spatial domain. Spatial- and frequency-domain features are enhanced separately and then fused. We also present a STEM noise calibration method for synthesizing more realistic STEM images. The parameters of scan noise, pointwise noise, and background noise are estimated through statistical analysis and fitting of real STEM images. Using these calibrated parameters, we construct a diverse dataset containing a wider range of atomic arrangements, imaging modes, and degradations. Random masks combine different atomic arrangements to simulate interfaces between multiple materials. Experimental results demonstrate that our method achieves state-of-the-art enhancement performance on synthetic datasets and real images of advanced materials while improving atomic localization accuracy.

cs.CV↗
arXiv2026

PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations

Vision--Language--Action (VLA) models generate robot actions from visual observations and language instructions, but their understanding of physical dynamics remains limited. Video world models can capture scene dynamics, yet generating future videos introduces high latency into the control loop. We introduce PDRVLA, whose lightweight Physical Dynamics Module extracts physical tokens from vision--language features to condition action generation without online video generation. The module learns through direct prediction of future RGB, depth, and optical-flow features, with auxiliary supervision through a frozen pretrained video world model. Training relies entirely on publicly available data, expanded through human-to-robot synthesis, subtask annotation, and embodied chain-of-thought (ECoT) annotation. Across four simulation benchmarks, PDRVLA achieves success rates of 99.3% on LIBERO, 89.9% on LIBERO-Plus, 93.6% on RoboTwin~2.0, and 81.5% on SimplerEnv. Ablations show improved future prediction with world-model supervision and evaluate how subtask conditioning, future object-pose supervision, and ECoT training affect action success.

cs.RO↗
arXiv2026

UnityBench-Mini3D: Evaluating LLM Agents for 3D Game Development in Unity

Game development stresses coding agents in ways general software engineering does not: correctness depends on how engine APIs, serialized scene state, and runtime behavior interact, not on source code alone. Existing benchmarks mainly target platforms such as Godot and heavily rely on LLM judges or coarse task-level outcomes. We present UnityBench-Mini3D, a benchmark of 77 project-level Unity tasks organized around seven expert-defined skill categories rather than tutorials or genres, with three difficulty tiers. Each task pairs a complete Unity project with natural-language requirements and is evaluated using 1,178 held-out, human-authored milestone tests covering scene structure, component configuration, and runtime behavior. This in-engine evaluation provides deterministic partial credit, localizes failures, and distinguishes partial implementations from complete solutions without relying on subjective judges. Across 21 models and 9 coding harnesses, the average pass rate reaches 80.9% on easy tasks but falls to 35.2% on hard tasks, with just 57.7% overall. UnityBench-Mini3D provides a reproducible, diagnostic testbed distinguishing systems across difficulty levels.

cs.SE↗
arXiv2026

TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social Environments

Robots must act under physical and social constraints while adapting to changes in their environment and human needs. Existing embodied-agent systems often focus on specified goals or abstract execution into symbolic actions, limiting evaluation in an evolving world. We introduce TARSphere, a robot-oriented full-stack framework for physically and socially grounded long-horizon embodied agents. TARSphere-Infra runs robot policies in a physics simulator with embodied NPCs and constrained dialogue. An event-driven finite-state machine (FSM) uses completed actions and delivered information to update task states and evaluate outcomes. TARSphere-Harness, its embodied brain harness, lets a frozen multimodal large language model (MLLM) act through multi-level executable skills, using observations, hierarchical memory, and execution feedback to revise its decisions. On this foundation, agent-in-the-loop curation aligns task semantics, causal dependencies, and scene geometry to construct TARSphere-Bench: 360 episodes across 30 indoor scenes. Five paired conditions assess perceptual robustness, disruption adaptation, task update, human coordination, and intent understanding for navigate-and-report and fetch-and-deliver tasks. Our evaluation of 8 MLLMs shows that general-purpose models can outperform embodied models, while task updates and human interaction remain difficult. Frequent collisions and falls further motivate evaluating execution safety alongside task success, even with a shared action interface and decision budget.

cs.RO↗
arXiv2026

InteractAgent: Agentic Human Motion Interaction with Memory-Augmented Reflection

Long-horizon human-scene interaction requires coordinated spatial reasoning, physical feasibility, and the ability to recover from compounding execution errors. Text-driven motion models are largely scene-agnostic and lack closed-loop correction, while geometry-conditioned approaches depend on paired scene-motion training or explicit 3D preprocessing that limits generalization to novel layouts and open-vocabulary goals. We present InteractAgent, an agentic multimodal LLM system for long-horizon human-scene interaction that plans, executes, observes, and reflects in a closed loop, using a single MLLM as both planner and evaluator and a frozen MotionLLM as the executor. InteractAgent integrates Online Observation for step-wise recovery decisions (continue/redo/replan) and Memory-Augmented Reflection for multi-round plan refinement with adaptive stopping. On UniHSI, InteractAgent improves task success and reduces penetration compared with training-free baselines. Code: https://github.com/AIGeeksGroup/InteractAgent.

cs.RO↗
arXiv2026

Manifold Prior Guided Deep Unfolding for Hyperspectral Image Reconstruction

Hyperspectral Image (HSI) reconstruction has made substantial progress with the deep unfolding framework by decomposing the problem into a projection module and a denoiser module. Nevertheless, existing methods still exhibit limitations of insufficient matching with HSI data. The issues lie in two aspects: 1) the data-consistency projection module applies a fixed gradient descent step while ignoring the spectral content adaptivity of HSI; 2) the denoiser module either suffers from high computational complexity or is limited to the local receptive field, failing to efficiently capture the global spatial-spectral dependencies. In this paper, we propose a deep unfolding framework centered on a two-stage manifold learning strategy to obtain a degradation-free structural prior, which is then strategically injected into both the projection and denoiser modules. In the data-consistency projection, a prior-guided spectral weight mechanism is introduced to adaptively modulate the weight of spectral band. In the denoiser module, a manifold-guided Mamba is proposed to achieve the structure-aware extraction of spatial-spectral features through prior modulation. Quantitative and visual comparisons on synthetic and real-world datasets show that the proposed method significantly outperforms existing state-of-the-art approaches, while maintaining the structural interpretability of the unfolding framework. The code is available at https://github.com/SqLiu67/MPG-DU.

cs.CV↗
arXiv2026

Ineffective Occupancy in Heterogeneous GPU Clusters: Prediction, Contribution Analysis, and Readiness-Aware Scheduling

In heterogeneous GPU clusters, GPUs are costly and scarce, yet placing a job on a machine does not mean that the job becomes ready to run. A large amount of allocated capacity can remain occupied for long periods without producing useful computation, further aggravating the waste of compute resources. On the Alibaba GPU cluster trace we find that, even when GPU capacity is sufficient and placement is complete, 52.70% of execution records never became ready. We term this phenomenon ineffective occupancy. Building on this diagnosis, and keeping the placement policy fixed, we propose Readiness-Aware Scheduling (RAS): it uses only nine attributes observable at admission time to predict readiness probability and reorders admission accordingly. Under a five-fold split grouped by machine, the gradient-boosted model reaches an ROC-AUC of 0.8187. A feature-group contribution analysis shows that the model's score is driven mainly by job profile, duration, and GPU model, with the request-size group contributing least, followed by the scheduling and concurrency group. In GPU terms, RAS reduces the ineffective-occupancy ratio (the share of cluster GPUs held under ineffective occupancy) from 43.64% to 27.15% and raises the effective coverage of large requests (eight GPUs or more) from 46.11% to 63.18%, without sacrificing capacity utilization. Relative to the upper bound obtained by ordering with ground-truth labels, RAS captures 65.1% of the attainable improvement, well above the interpretable rules built only from the contribution-analysis conclusions; the gain therefore comes mainly from predicting with all nine attributes jointly rather than from combining rules built on any single attribute group. These results indicate that readiness status is a measurable, predictable, and optimizable dimension of admission decisions in heterogeneous GPU clusters.

cs.DC↗
arXiv2026

World Action Critic (WAC) Model: Joint Foresight and Criticism for Policy Learning

World models have shown considerable promise in sequential decision-making by anticipating the consequences of actions and providing predictive foresight for policy learning. However, learning to reason about counterfactual actions remains challenging, since predictive objectives may underemphasize information critical for action evaluation, while unexecuted actions lack observed futures for direct supervision. To address these challenges, we propose the World Action Critic (WAC) Model, which proposes multimodal action candidates and jointly learns candidate-conditioned future prediction and consequence evaluation. In our proposed framework, critic supervision directly shapes intermediate predictive representations during rollout, encouraging them to retain information needed to distinguish action quality. Observed futures supervise prediction for executed actions, while external task-quality targets provide evaluation signals for counterfactual candidates. Through joint optimization, critic signals from these candidates propagate into predictive representations without requiring their observed futures. Our theoretical analysis characterizes critic-guided feature retention and bounds candidate-selection regret. Applied to autonomous driving, WAC achieves 93.4 PDMS on NAVSIM-v1, demonstrating the effectiveness of jointly learning predictive foresight and consequence evaluation.

cs.AI↗
arXiv2026

Affordance-Aware Interactive Decision-Making and Execution for Ambiguous Instructions

Enabling robots to explore and act in unfamiliar environments under ambiguous human instructions by interactively identifying task-relevant objects (e.g., cups or beverages for "I'm thirsty") remains challenging for existing vision-language model (VLM)-based methods, due to inefficient reasoning and limited environmental interaction. These factors hinder real-time task planning and execution. To address this, we propose Affordance-Aware Interactive Decision-Making and Execution for Ambiguous Instructions (AIDE), a human-inspired dual-stream framework that integrates interactive exploration with vision-language reasoning: mirroring the complementary deliberative and intuitive processes of human cognition, Multi-Stage Inference (MSI) serves as the decision-making stream and Accelerated Decision-Making (ADM) as the execution stream, enabling zero-shot affordance analysis and interpretation of ambiguous instructions. Experiments in simulation and real-world settings show that AIDE achieves over 80% task planning success rate and more than 95% closed-loop continuous execution accuracy at 10 Hz, surpassing existing VLM-based methods in diverse open scenarios.

cs.RO↗
arXiv2026

RM-WAM: Reliable Memory-Augmented World Action Models for Long-Horizon Robot Manipulation

World Action Models (WAMs) improve robotic manipulation by jointly predicting future visual states and actions, yet most existing methods make decisions primarily from the current observation. This limits their reliability in long-horizon tasks, where visually similar states may correspond to different interaction histories and task stages. Moreover, directly incorporating historical observations can introduce stale, stage-mismatched, or error-contaminated information, leading to inaccurate world prediction and accumulated closed-loop errors. We propose RM-WAM, a reliable memory-augmented World Action Model that treats memory as a context-dependent predictive resource rather than universally useful historical context. RM-WAM represents history using complementary short-term visual state memory and latent event memory, which respectively capture recent interaction dynamics and longer-term state transitions. A counterfactual memory learning scheme varies the accessible history while keeping the current inputs and prediction targets fixed, enabling the model to learn a context-dependent utility estimate for historical memory. Based on this estimate, a utility-aware gating mechanism adaptively controls memory injection into the world-action backbone. Experiments demonstrate that RM-WAM highlights the importance of reliable and selective memory utilization in long-horizon tasks.

cs.RO↗
arXiv2026

ZoomCap: Evidence-Routed Visual Zooming for Recoverable Visual Information Transfer

Detailed captions transfer visual information into vision-language model supervision, but quality is easily conflated with length: long single-pass outputs still miss small objects, text, and spatial relations or promote uncertain observations into confident claims. We formulate fine-grained captioning as recoverable visual information transfer, in which a teacher acquires missing evidence under a finite observation budget and converts it into reusable supervision. ZoomCap uses a draft caption to route conditional knowledge retrieval, OCR-sensitive reading, coverage-gap recovery, and adaptive visual zooming, allocating deeper crops only to the most complex eligible region, deriving spatial relations from grounded geometry, and fusing non-duplicate evidence. We also introduce FactCycle, which scores caption-derived atomic claims on two blind anchors--source image (truthfulness) and reconstruction (realizability)--separating fabrication from generator loss claim by claim; human and multi-judge comparisons complement this fact-level view. Against Qwen3.5-397B-A17B captions, ZoomCap achieves an 81.9% human tie-adjusted preference score with a teacher over 10x smaller, and FactCycle counts 52.4 transferred claims per image vs. 31.5 for ScaleCap. The evidence also transfers: in continual pre-training, ZoomCap-450K improves the 11-benchmark average by 1.31-2.38 points over the strongest competing corpus and Ling-Tiny-VL-v3 gains 1.4 points on a production document/OCR monitor suite (1.5 on perception) at 60B tokens; in medical pre-training, 2.64M ZoomCap-captioned images improve five VQA benchmarks by 3.04 points (vs. 1.76 for same-image Gemini-3.1-Pro captions).

cs.CV↗
arXiv2026

SkillMarker: A KGW-Based Watermarking Method for Agent Skill Ownership Verification

Agent skills encode operational procedures and domain expertise as reusable knowledge assets, yet their distribution in plaintext exposes them to unauthorized copying and leakage. However, practical mechanisms for skill ownership verification and leakage tracing remain underdeveloped. Watermark embedding before distribution offers a potential solution, but structured, compact skill instructions impose stringent requirements for factual and procedural accuracy. To address these challenges, we introduce SkillMarker, a KGW-based watermarking method that jointly targets functionality preservation and strong watermark embedding. SkillMarker unifies quality-constrained rewriting with signal-oriented candidate search. The former separates essential knowledge from flexible expression to preserve instruction fidelity; the latter explores alternative expressions to obtain a stronger watermark signal under preservation constraints. We release an open-source toolkit that integrates watermark embedding, quality auditing, and detection to support the protection of agent skills. SkillMarker targets ownership verification and leakage tracing in distributed skill ecosystems, where content may be copied in whole or in part, or incorporated into other skills. Across four datasets, SkillMarker negligibly affects downstream task performance on average. In our copy-paste tests, window-based detection consistently identifies copied passages containing as few as 1,536 consecutive tokens. Even after merging with one or three background skills, SkillMarker retains watermark detection rates of 72.2--77.8\%.

cs.CR↗
arXiv2026

Atomic-WAM: Execution-Grounded Atomic-Motion Reasoning for World Action Models

World Action Models (WAMs) typically adapt video foundation models to robotic control through joint video and action prediction. However, task-level instructions may leave the next physical interaction underspecified, requiring semantic grounding and state-dependent motion selection. To this end, we introduce Atomic-WAM, an understand-then-execute framework that connects task reasoning to action generation through atomic-motion descriptions. Specifically, a vision-language reasoner describes the next local motion from the task instruction and visual context, guiding a WAM to predict future video and an action chunk. Furthermore, an automatic annotation pipeline curates temporally aligned, contact-enriched supervision from multi-view DROID demonstrations. To make atomic-motion predictions executable by the WAM, we post-train the reasoner with Execution-Grounded GRPO (EG-GRPO). With the WAM fixed, EG-GRPO adapts the reasoner through rewards for motion appropriateness and execution success. Finally, Atomic-WAM achieves 39.6% task success under default instructions and 31.6% under vague instructions in a real-to-sim evaluation on RoboLab. Project page: https://ssj9596.github.io/Atomic-WAM-project

cs.RO↗
arXiv2026

AutoRSR: Automated Real-to-Sim-to-Real Data Generation for Scalable Robot Learning

Scaling embodied robot learning requires large and diverse manipulation datasets, but collecting real-world demonstrations remains expensive in human effort and robot time. Simulation provides a scalable alternative, yet converting a real workspace into a robot-learning-ready environment still requires substantial manual effort for object extraction, asset processing, coordinate calibration, and trajectory specification. We present AutoRSR, a highly automated real-to-sim-to real framework that converts scanned real-world workspaces into reusable embodied simulation assets and scalable manipulation data. Given multi-view observations, robot states, and open-vocabulary semantic queries, AutoRSR reconstructs the workspace, extracts individual object assets, and automatically registers the reconstructed scene to the robot base frame at metric scale. Starting from a small number of reference motions, AutoRSR then generates executable manipulation trajectories and scales them through trajectory interpolation and domain randomization. After workspace scanning and semantic queries, the remaining pipeline requires no manual object annotation, scene-to-robot calibration, or robot teleoperation. Across three real-world manipulation tasks, policies trained on 1K automatically generated episodes per task achieve 73.3--86.7% real-world success rates, comparable to pi 0.5 trained from 150 human teleoperated demonstrations, while reducing the human time required for data generation by an order of magnitude. These results demonstrate that real-world observations can be converted into metrically aligned, executable, and scalable simulation data with substantially reduced human effort.

cs.RO↗
arXiv2026

Learning native eddy-resolving ocean dynamics outperforms coarse emulation and downscaling

Eddy-resolving ocean simulations are essential for representing mesoscale dynamics that regulate ocean circulation, energy transport, and climate variability. However, they are costly to generate, store, and share. Artificial intelligence (AI) may help, but it remains unclear whether high-resolution ocean fields can be faithfully represented by evolving ocean states at coarse resolution followed by spatial downscaling, or whether dynamical evolution needs to be learned directly from native high-resolution simulations. We show that directly learning dynamical evolution on a 0.1$^\circ$ grid provides a more faithful representation of mesoscale variability than evolving ocean states on a 1$^\circ$ grid followed by downscaling reconstruction. We developed a full-depth 1$^\circ$ emulator and a native 0.1$^\circ$ sea surface temperature (SST) emulator, both of which remained stable over long-term free-running simulations without significant drift, providing reliable baselines for this comparison. Direct learning at 0.1$^\circ$ preserves mesoscale energy, spectral characteristics, and frontal gradients, whereas the coarse-resolution evolution followed by downscaling approach recovers large-scale patterns but fails to reproduce key mesoscale signatures. We show that the learned mesoscale statistics remain consistent across diverse hardware platforms, supporting the reproducibility and portability of AI-based ocean simulations. These results suggest that downscaling of low-resolution large-scale states alone is insufficient to represent ocean mesoscale variability, supporting the need to learn dynamical evolution directly in a high-resolution state space. These findings highlight the continued importance of high-resolution physics ocean models as a foundation for AI-based ocean simulation, providing dynamically resolved training data for physically faithful high-resolution emulators.

physics.ao-ph↗
arXiv2026

The throttling refrigeration system for high cooling power recovery of the PandaX-xT cryogenic distillation system for radon removal

The PandaX-xT cryogenic distillation system for removing radon (Rn) is designed to operate at a high liquid flow rate of 856 kg/h, equal to 5 liters per minute (LPM), to meet the required Rn concentration. However, it is challenging to maintain a continuous supply of cooling power (20 kW) for the Rn-removal system. This work proposes a throttling refrigeration system to address the power supply problem for the next-generation PandaX-xT dark matter detection experiment. The system utilizes carbon tetrafluoride (R14) refrigerant to recover power based on heat pump principles. The cooling power from the liquid xenon (LXe) in the reboiler at 178 K is transferred to liquefy the gaseous xenon (GXe) in the cryostat through the phase change circulation of R14, enabling xenon (Xe) to return to the detector in the same state as when it was extracted. Consequently, a stable circulation for liquid distillation is established. Given the high cost of Xe, a research and development (R&D) experiment has been conducted to validate the feasibility of this high-cooling-recovery system by using ethanol as a substitute for Xe. Experimental results indicate that the proposed system could achieve a cooling power recovery of 25 kW, with a recovery efficiency of 99%+-0.5%. Additionally, process simulation and optimization of the throttling refrigeration cycle are performed using Aspen Hysys. The deviation between the simulation and experimental results is found to be <3.64%. These findings demonstrate that implementing Rn-removal distillation at a high flow rate is feasible, eliminating the need for liquid nitrogen or cryocoolers. This system could save approximately 3650 m3 of liquid nitrogen per year or reduce power consumption by 230 kW.

physics.ins-det↗
arXiv2026

DASC: Decay-Aware State Compression for Hybrid Linear-Attention Serving

Hybrid linear-attention architectures have recently scaled to large open-weight models, offering quality competitive with full attention while substantially reducing key/value (KV) cache growth. However, their in-place recurrent-state updates complicate cache management: prefix reuse requires state checkpoints alongside full-attention KV, while storing state checkpoints in full increases memory pressure, leading to more evictions and repeated prefill. By analyzing the decay structure of Gated DeltaNet (GDN) and Kimi Delta Attention (KDA), we find that different heads and channels retain prefix information over markedly different timescales, which we term \emph{retention horizons}. This variation suggests substantial compression potential in persistent state checkpoints. Building on this observation, we introduce \emph{Decay-Aware State Compression} (DASC), which derives retention horizons from model weights, selects long-horizon state units, and packs them into a ragged state checkpoint layout. To integrate efficiently with tensor-parallel inference engines, DASC furtherly balances compressed state checkpoints across TP ranks. On reuse, DASC either zero-fills omitted units or refreshes them from a bounded suffix with additional compute cost. Across retrieval and end-to-end reasoning benchmarks on Kimi-Linear, conservative DASC configurations remain close to full caching while compressing KDA recurrent state checkpoints by $2.63\times$. Under fixed state checkpoint memory budgets, the resulting capacity gains reduce mean Time to First Token (TTFT) by 42.6\% and improve input throughput by 68.4\%. At larger compression ratio, suffix refresh recovers much of the accuracy lost to more aggressive omission, at the cost of additional replay computation. Qwen with GDN exhibits a similar quality--efficiency trend, showing that DASC extends from channel-wise KDA to head-wise GDN.

cs.LG↗