Search arXiv⌕ Search

arXiv subjects

Yang Wang

Publications and source records attributed to Yang Wang.

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

One Rollout Is All You Get: Fully Test-Time Adaptation for GUI Agents

GUI agents are deployed with frozen weights and discard everything they experience on the job. Existing ways to update an agent's weights assume something deployment withholds: ground truth, rollouts beyond the single attempt (retries, samples, practice runs), or a learning phase other than deployment. Because GUI actions can be irreversible, a deployed agent gets one attempt per task occurrence, in arrival order, and every attempt counts. No ground truth is available at any point. We define fully test-time adaptation for GUI agents by these constraints and pair it with a minimal weight-space method, SOLO. Auxiliary models read each episode: a judge selects the episodes it deems successful, and a proposer-verifier pair relabels a failed episode's prefix with the subtask that prefix completed. Admitted episodes enter a short sliding window, and each admission updates a small adapter by top-K self-distillation on the agent's own predictions, provided the window holds a judged success. On recurring task streams built from WebArena, VisualWebArena and MobileWorld, SOLO improves on the frozen agent with both UI-TARS-7B and Qwen3-VL-8B, by three to six points of success rate, and exceeds two in-setting memory methods on the web streams.

cs.LG↗
arXiv2026

Remember by Asking: Retrieval-Induced Memory Evolution for LLM Agents

Long-term memory is essential for language agents to maintain coherent and effective behavior over extended, multi-session interactions. Existing memory systems mainly use retrieval at read time, while write-time memory formation still relies on direct extraction or compression. However, when future information needs are unknown, compressing an entire interaction in one pass can overlook locally important details that may matter later. To this end, we introduce RIME, a retrieval-induced memory framework that shifts memory construction from monolithic compression toward evidence-centered integration. RIME uses generic self-questions to retrieve focused dialogue evidence and grounds memory formation in both the retrieved evidence and relevant historical memories, which are jointly reconciled into an evolving memory bank with temporal and provenance information. At inference time, compressed memory serves as the primary rather than the sole source of evidence: when it cannot support an answer, RIME retrieves relevant source dialogue together with its local context to recover information omitted during memory formation, without resorting to full-history processing. Extensive experiments on LoCoMo with Qwen3-235B-A22B and GPT-5.6 Sol show that RIME consistently achieves the best performance across all three quality metrics among the compared methods, while requiring substantially fewer query-time LLM tokens.

cs.CL↗
arXiv2026

SymbolicArena: A Unified Infrastructure for Benchmark Distillation and Dynamic Evaluation in Symbolic Regression

Symbolic regression (SR) seeks concise and interpretable mathematical expressions from data for scientific equation discovery. Existing SR benchmarks face a tradeoff between evaluation cost and benchmark validity. Repeated evaluation of large task pools is expensive, and compact benchmarks lack systematic evidence of preserved task diversity and algorithm discriminability. SymbolicArena provides a unified infrastructure for benchmark distillation and dynamic evaluation. The framework standardizes 664 heterogeneous tasks with executable ground truth expressions and distills the Full Task Set into Core50, a validated benchmark of 50 tasks. The distillation process preserves task coverage and algorithm discrimination under explicit balance constraints. SymbolicArena applies a unified execution protocol to heterogeneous SR algorithms and produces comparable outputs and search trajectories. Multi Axis Evaluation characterizes numerical quality, symbolic quality, and search behavior. Core50 reduces evaluation workload by 92.5% and maintains agreement with Full Task Set evaluations. Experiments show that SymbolicArena achieves 72.6% to 86.7% lower approximation error than alternative selectors, further supporting its fidelity to the Full Task Set. Evaluation reveals a substantial gap between numerical fitting and symbolic recovery across current SR methods, suggesting that reliable equation recovery remains an open challenge.

cs.LG↗
arXiv2026

ShellfishNet: A Domain-Specific Benchmark for Visual Recognition of Marine Molluscs

Reliable visual recognition of shellfish is important for ecological monitoring and aquaculture, but real-world imagery exhibits substantial variation in acquisition conditions, appearance, and posture. Existing marine vision resources cover diverse taxa and imaging settings, but fine-grained mollusc recognition remains underexplored in a group-disjoint benchmark that jointly evaluates class prediction, representation-space retrieval, and condition sensitivity. To address this gap, we introduce ShellfishNet, a provenance-aware, group-disjoint benchmark of 40,130 images from 100 marine mollusc classes, organized into 30,879 acquisition or observation groups. Rather than treating recognition as a single clean-accuracy leaderboard, ShellfishNet evaluates model configurations across classification, image-to-image retrieval, taxonomic error structure, and controlled condition sensitivity. We benchmark representative convolutional, transformer-based, hybrid and efficient, self-supervised, and vision-language models on classification and retrieval. Separately, we evaluate open image-captioning and vision-language models on the captioning subset and assess selected fixed checkpoints under controlled generic and marine-inspired synthetic transformations. Across the benchmark, image-level and class-aware classification largely agree, whereas retrieval provides complementary model rankings and taxonomy projection reveals that a non-negligible subset of exact-class errors remains correct at higher taxonomic ranks. Selected fixed checkpoints with similar clean accuracy can also exhibit distinct sensitivity profiles under controlled transformations. These results show that model comparison on shellfish imagery should account for both task and image condition, rather than rely on a single clean classification score. The dataset and evaluation protocols will be publicly released.

cs.CV↗
arXiv2026

ProcVLM: Learning Procedure-Grounded Progress Rewards for Robotic Manipulation

Long-horizon robotic manipulation requires dense feedback that reflects how a task advances through its procedural stages, not merely whether the final outcome is successful. Existing reward models often rely on trajectory-level success labels or time-based interpolation, which can conflate elapsed time with true task progress and therefore fail to capture unfinished steps, stagnation, and failure states. We present ProcVLM, a progress-aware vision-language model that learns procedure-grounded progress as a dense reward signal for manipulation. Rather than deriving progress from terminal outcomes or temporal proxies, ProcVLM grounds progress estimation in procedural structure and intra-stage visual change, and further adopts a reasoning-before-estimation paradigm that infers the remaining atomic actions before estimating task progress. Specifically, we construct this supervision by synthesizing frame-level subtask-semantic annotations, assigning progress budgets according to subtask structure, and distributing each budget based on intra-subtask visual change. To train ProcVLM at scale, we build a standardized procedural supervision synthesis pipeline and construct ProcCorpus-60M from 30 embodied datasets with 60M annotated frames, from which we derive ProcVQA for procedure-aware pretraining, with progress estimation as the central task alongside action segmentation and future planning. Experiments on ProcVQA and reward-model benchmarks show that ProcVLM improves embodied procedural reasoning and yields more discriminative trajectory-internal progress estimates than representative baselines, supporting its use as a dense reward model for downstream reward-guided policy optimization. Project page: https://procvlm.github.io/

cs.RO↗
arXiv2026

Finite-Time Doppler Covariance Spectroscopy in the AdS/CFT correspondence

From a holographic perspective, we formulate an exact finite-duration covariance framework that resolves chirality in the AdS/CFT correspondence. In a rotating BTZ black hole spacetime, relative motion between two boundary controls converts sum-frequency matching on the compact boundary cylinder into discrete velocity resonances indexed by the angular momentum of the scalar field. Because the complete real-control covariance also contains difference-frequency sectors, a two-setting phase cycle with a common switching window cancels them exactly at finite duration and reconstructs the resonant pair contribution from complete positive-semidefinite covariance matrices. For an isolated angular mode, reconstructed pair responses at opposite resonance velocities sample the boundary spectrum at opposite angular momentum and a common control-selected frequency. Their normalized contrast cancels reflection-symmetric control factors and the state-independent operator normalization, yielding a finite-linewidth estimator of the reflected-mode BTZ contrast. Under the stated smoothness and moment conditions, common centered even narrowing windows give a bias beginning at quadratic order in linewidth. A real-projection leakage bound controls contamination from neighboring modes. Long-pulse calculations with the exact BTZ spectrum verify phase-cycle reconstruction and quadratic linewidth scaling. Bath-weighted calculations with exact truncated transfer amplitudes establish convergence to this baseline. Under the default controls, complete-sum contrasts for the first two nonzero angular modes reconstruct the reference-state rotation parameter at fixed mean inverse temperature with subpercent deterministic bias in the untruncated Gaussian limit. The inferred rotation parameter fixes the horizon parameters of the rotating BTZ black hole once the mean inverse temperature and AdS radius are known.

hep-th↗
arXiv2026

MicroHookACT: Monocular Microscopic Vision Guided Visuomotor Policy for Flexible Microelectrode Hooking

Automated needle-loop hooking is a critical step in flexible microelectrode (FME) implantation. This paper presents MicroHookACT, an imitation learning-based visuomotor policy for automated 3D hooking under monocular microscopic vision. First, a unidirectional hooking strategy exploits defocus cues and optical-axis guidance to enable palpation-free precise alignment and contact-rich threading. Second, an action-supervised object attention module built on a frozen ViT backbone learns to focus on the micro-needle tip and micro-loop directly from human demonstrations, without requiring manual visual annotations for training. Third, attention-centered global coarse and local fine features are dynamically weighted according to predicted action progress, enabling a single ACT policy to adapt to changing defocus blur and visual requirements throughout the operation. In the experiments, visuomotor policies were trained on 60 human demonstrations and evaluated under five setups with varying difficulties. Our MicroHookACT framework achieved the highest overall success rate of 96.7\% with an average execution time of 11.5 s. These results demonstrate the potential of visuomotor policy learning for micron-level control under varying operating conditions.

cs.RO↗
arXiv2026

GigaBrain-WBC-0.5: A Behavior World Model for Robust Humanoid Whole-Body Tracking with Environment Interaction

General-purpose motion trackers enable humanoid robots to follow diverse whole-body motions while maintaining balance, but are trained only on flat ground, failing to exploit bipedal mobility over complex terrain. Cross-terrain controllers, meanwhile, are task-specific or accept only low-dimensional locomotion commands. We introduce InterTrack, the first behavior world model (BWM) for robust whole-body tracking with environment interaction. Its Transformer jointly predicts the next action, state, and behavior distribution, learning environment-conditioned dynamics. To scale interaction training data, an automatic annotation pipeline reconstructs 3D support geometry from retargeted motions. At deployment, the policy handles commands implausible in the current environment in a "best-effort" manner. Quantitatively, InterTrack achieves an 81.3% success rate on terrain interaction (4.3 times the best evaluated baseline) and a 99.3% fall-recovery rate, while also improving free-space tracking and outperforming three leading tracking baselines across all of these regimes. To our knowledge, we provide the first demonstration of real-time cross-terrain whole-body teleoperation on a humanoid robot, alongside object interaction, stable responses to missing supports, and robust recovery from falls.

cs.RO↗
arXiv2026

Mechanism of charge transfer and electrostatic field fluctuations in complex metallic alloys

Complex metallic alloys exhibit rich disorder-driven electronic, magnetic, and vibrational behavior arising from strong chemical disorder, leading to unconventional structure-property relationships with applications in extreme-environment materials, catalysis, spintronics, and thermoelectrics. Despite their technological relevance, the microscopic nature of charge redistribution and electrostatic fluctuations in chemically disordered alloys remains incompletely understood. Here, we develop a theoretical framework that uncovers universal statistical trends of disorder-driven charge transfer and Madelung-field fluctuations. Our analytical formalism demonstrates that local charge transfer and electrostatic potentials exhibit Gaussian-like statistics and universal linear charge-potential (qV) correlations emerging directly from the underlying disorder landscape. We identify the physical origin of these correlations in the interplay between electronic screening and impurity scattering and show how their statistical properties depend on carrier density, disorder strength, and compositional complexity. We further derive scaling relations governing qV trends across binary and multicomponent alloys, including high-entropy materials. Large-supercell density-functional theory (DFT) calculations show good quantitative agreement with the predicted statistical behavior for representative metallic alloys. Our results provide a computationally efficient framework for incorporating disorder-driven electrostatic fluctuations into effective-medium electronic-structure theories and establish a statistical-physics foundation for the predictive design of complex alloys.

cond-mat.mtrl-sci↗
arXiv2026

Partial GFlowNet: Accelerating Convergence in Large State Spaces via Strategic Partitioning

Generative Flow Networks (GFlowNets) have shown promising potential to generate high-scoring candidates with probability proportional to their rewards. As existing GFlowNets freely explore in state space, they encounter significant convergence challenges when scaling to large state spaces. Addressing this issue, this paper proposes to restrict the exploration of actor. A planner is introduced to partition the entire state space into overlapping partial state spaces. Given their limited size, these partial state spaces allow the actor to efficiently identify subregions with higher rewards. A heuristic strategy is introduced to switch partial regions thus preventing the actor from wasting time exploring fully explored or low-reward partial regions. By iteratively exploring these partial state spaces, the actor learns to converge towards the high-reward subregions within the entire state space. Experiments on several widely used datasets demonstrate that \modelname converges faster than existing works on large state spaces. Furthermore, \modelname not only generates candidates with higher rewards but also significantly improves their diversity.

cs.LG↗
arXiv2026

The Fusion Frame Phase Retrieval

The phase retrieval problem involves reconstructing a function or signal solely from the magnitude of linear measurements. Most theoretical analyses of phase retrieval algorithms rely on i.i.d. Gaussian random measurements or sub-Gaussian random measurements. In this paper, our focus is on the fusion frame phase retrieval problem, where the sampling matrices are i.i.d. rank-$r$ orthogonal projections drawn from the Haar measure. We present concentration inequalities for functions on the set of rank-$r$ orthogonal projection matrices. These inequalities are crucial for the theoretical analysis of the fusion frame phase retrieval problem. Based on these inequalities, we demonstrate that gradient descent, combined with a two-stage initialization, achieves linear convergence to the target signal up to a global phase with a measurement complexity of $O(d\log^2 d)$ when the rank $r = O(1)$. We verify this convergence through numerical results.

cs.IT↗
arXiv2026

Augur: Predicting View Serializability Violations in Relational Data Store Applications

Data stores are widely used because they provide persistence, scalability, and fault tolerance with a simple interface. However, most data store applications configure the data store to use weak isolation to achieve scalable performance, resulting in sporadic unserializable executions that are incorrect or fail. Prior work uses dynamic predictive analysis to infer violations from execution traces, but it cannot handle relational (i.e., SQL) queries with complex predicates, and it predicts executions that do not violate View Serializability. This paper introduces Augur, the first dynamic predictive program analysis that (1) supports data store applications with complex relational queries and (2) reports only executions that violate View Serializability. The evaluation demonstrates that Augur finds feasible, unserializable executions in the widely used OLTP-Bench programs and in the widely used e-commerce application Spree.

cs.PL↗
arXiv2026

Stop Before You Fail: Operational Capability Boundaries for Mitigating Unproductive Reasoning in Large Reasoning Models

Current answering paradigms for Large Reasoning Models (LRMs) often fail to account for the fact that some questions may lie beyond the model's operational capability boundary, leading to long but unproductive reasoning. In this paper, we study whether LRMs expose early signals predictive of such cases, and whether these signals can be used to mitigate unproductive reasoning. In black-box settings, we find that reasoning expressions contain failure-predictive signals. In white-box settings, we show that the hidden states of the last input token contain information that is predictive of whether a question will not be solved correctly under our evaluation setup. Building on these observations, we propose two test-time monitoring strategies: reasoning expression monitoring and hidden states monitoring, that reduce token usage by 62.7-93.6%, substantially improving efficiency and reliability while largely preserving accuracy.

cs.AI↗
arXiv2026

Propagation dynamics of an acid-mediated invasion model with degenerate tumor diffusion

We investigate traveling wave fronts for an acid-mediated tumor invasion model with density-dependent degenerate diffusion. The model is a partially diffusive PDE--ODE system of Gatenby--Gawlinski type, in which the tumor diffusion coefficient $D(U)$ is allowed to be a general decreasing function satisfying $D(1)=0$. This degeneracy causes the traveling wave equation for the tumor component to lose uniform ellipticity near the healthy state, and hence standard arguments for nondegenerate reaction diffusion systems are not directly applicable. To overcome this difficulty, we introduce a nonlinear change of variables which removes the degeneracy from the highest-order term of the tumor equation. For each fixed admissible tumor profile, the acid profile is represented by a Green kernel, while the healthy-tissue profile is obtained from an explicit integral formula. The transformed tumor profile is then constructed as the stationary limit of a uniformly parabolic auxiliary problem. By combining comparison principles, local Schauder estimates, carefully chosen super- and sub-solutions, and the Schauder fixed-point theorem, we prove the existence of traveling wave fronts for every wave speed $θ\ge 2\sqrt{rD(0)}$. The resulting wave connects the tumor-dominant state $(0,1,1)$ at $z=-\infty$ to the healthy state $(1,0,0)$ at $z=+\infty$. We further establish strict pointwise bounds, monotonicity of all wave components, and one-sided exponential asymptotic estimates in both the transformed variable and the original traveling-wave variable.

math.AP↗
arXiv2026

RobustGS: Unified Boosting of Feedforward 3D Gaussian Splatting under Low-Quality Conditions

Feedforward 3D Gaussian Splatting (3DGS) overcomes the limitations of optimization-based 3DGS by enabling fast and high-quality reconstruction without the need for per-scene optimization. However, existing feedforward approaches typically assume that input multi-view images are clean and high-quality. In real-world scenarios, images are often captured under challenging conditions such as noise, low light, or rain, resulting in inaccurate geometry and degraded 3D reconstruction. To address these challenges, we propose a general and efficient multi-view feature enhancement module, RobustGS, which substantially improves the robustness of feedforward 3DGS methods under various adverse imaging conditions, enabling high-quality 3D reconstruction. The RobustGS module can be seamlessly integrated into existing pretrained pipelines in a plug-and-play manner to enhance reconstruction robustness. Specifically, we introduce a novel component, Generalized Degradation Learner, designed to extract generic representations and distributions of multiple degradations from multi-view inputs, thereby enhancing degradation-awareness and improving the overall quality of 3D reconstruction. In addition, we propose a novel semantic-aware state-space model. It first leverages the extracted degradation representations to enhance corrupted inputs in the feature space. Then, it employs a semantic-aware strategy to aggregate semantically similar information across different views, enabling the extraction of fine-grained cross-view correspondences and further improving the quality of 3D representations. Extensive experiments demonstrate that our approach, when integrated into existing methods in a plug-and-play manner, consistently achieves state-of-the-art reconstruction quality across various types of degradations.

cs.CV↗
arXiv2026

Explainable Multimodal Depression Recognition in Clinical Interviews via PHQ-Aligned Symptom Summarization

Recent advances in multimodal depression recognition for clinical interviews (MDRC) have demonstrated the potential of AI systems by integrating textual, acoustic, and facial cues. However, existing methods pay limited attention to interpretability, thereby constraining reproducibility and clinician review. To address this, we introduce Explain-MDRC, an explainable MDRC framework that mirrors clinical workflows by generating structured symptom summaries from text and integrating them with nonverbal cues for recognition. Specifically, we construct Explain-DAIC, a dataset based on DAIC-WOZ and enriched with PHQ-8-aligned summary annotations, providing a foundation for developing models with built-in interpretability. We further propose PhqCML, a model that combines PHQ-8-aligned symptom summarization with PHQ-aware contrastive learning and summary-informed multimodal fusion. Automated metrics and expert evaluations show that Explain-MDRC improves recognition performance and provides more interpretable, clinician-readable intermediate evidence, suggesting a promising direction for transparent AI-assisted depression recognition research.

cs.CL↗
arXiv2026

Deterministic DTFT Interpolation for Joint Frequency and Chirp-Rate Estimation: Cell-Uniform Efficiency and Threshold Analysis

Joint frequency and chirp-rate estimation for a noisy chirp signal arises in radar, sonar, and burst satellite communications. Conventional estimators combine a coarse grid search with fine interpolation; accuracy degrades at the edges of the residual cell (the edge effect) and below the breakdown SNR (the threshold effect). We present a deterministic two-stage estimator that controls both failure modes uniformly over the residual cell. The estimator combines a time-centered, zero-padded dechirp-FFT acquisition bank with alternating selectable-$p$ amplitude-interpolation refinements on DTFT samples at fractional bins; in the centered frame, the frequency-chirp-rate cross-term of the Fisher information vanishes. The paper derives a mean-squared-error and threshold characterization over the full SNR range, in closed form except for one calibrated scalar (an effective cell count), to our knowledge the first for the joint problem: the breakdown threshold is governed by the cell count, and its cell-position dependence is dominated by the scalloping loss of the coarse FFT, which the padding bounds at 0.4 dB. An asymptotic uniformity analysis over the cell, including its corners, gives fixed-point variance ratios of $1.003$ and $0.998$, analytically free of the residual. A closed-form bias analysis under a cubic phase mismatch shows the centered chirp-rate estimate is insensitive to first order. Monte Carlo experiments at $N=256$ (validated at $N=32$-$512$) measure frequency- and chirp-rate-axis efficiencies with median $1.03$ and worst case $1.07$ over $144$ cell positions at $-5$ dB. Threshold predictions hold within $1.0$ dB on four configurations not used in the calibration. The dechirp-FFT bank is fully parallel, and each of the four refinement iterations evaluates three DTFT samples per axis; under fixed operating conditions, per-estimate latency is constant at $O(N\log N)$ cost.

eess.SP↗
arXiv2026

When Tool-Backed Skill Retrieval Fails: Source-Style Collapse in Executable Capability Retrieval

Large-scale agents increasingly rely on retrieval to access external capabilities. We study this retrieval gate in structured tools and APIs, a measurable class of tool-backed executable skills that must be surfaced before an agent can plan, incorporate, or act. In this setting the retrieval layer can silently fail even when the capability corpus is fixed: on ToolRet, a retriever fine-tuned on one source-specific slice collapses on another source-specific slice of the same benchmark, with FT-1100 despite its higher lexical overlap with the gold tools. We call this failure mode source-style collapse. Query-side TF-IDF fingerprints flag source styles on which the fine-tuned retriever is likely to fail better than semantic or length-based proxies, giving a cheap signal for mismatch over a fixed tool corpus. We propose ToolScout, a source-aware routing method that uses this signal as a routing guard: on the mixed 4,996-query stream, TF-IDF-based routing raises coverage from 22.3% to 86.1%, and across five collapsed sources 20 matched examples raise the coverage-weighted global top-1 proxy from 1.3% to 53.9%. The same failure and routing behaviors persist when tools are rerendered as executable skill cards, which rules out raw API-schema format as the sole cause.

cs.LG↗