Searcharxiv⌕ Search

arXiv subjects

Yi Zhang

Publications and source records attributed to Yi Zhang.

At least 19 recordsLinked to original sources
arXiv2026

GUITestScape: Towards Open-set Evaluation on Exploratory GUI Testing

Exploratory GUI testing is a particularly demanding setting for MLLM agents: without predefined test scripts, an agent must autonomously navigate an application and discover defects through its own interaction. However, current evaluation falls short on two fronts. First, existing benchmarks focus almost exclusively on interaction defects, leaving display defects outside the evaluation frame. Second, evaluation protocols are bound to predefined defect annotations, collapsing the testing process into a single end-state judgment that conflates qualitatively distinct failure modes. To address these challenges, we present GUITestScape, an interactive benchmark covering 61 real-world Android applications and 508 preset defects spanning interaction and display types, and introduce GUIJudge, an open-set evaluator that decomposes an agent's testing trajectory into independently diagnosable capabilities. Experimental results demonstrate that GUIJudge achieves reliable process-aware evaluation beyond predefined annotations, substantially outperforming all baselines. Benchmarking on GUITestScape further reveals that detection remains the critical bottleneck for existing models across both defect types, and that integrating GUIJudge's verifiers into existing agents significantly boosts their detection performance without retraining.

cs.SE↗
arXiv2026

InfantNet: A Large Scale Dataset for Infant Body Pose and Shape Estimation

Infant pose and shape estimation is essential for applications in childcare, developmental monitoring, and medical diagnosis. However, existing methods and datasets are largely designed for adults, and direct transfer to infants fails due to substantial differences in body proportions, articulation limits, and frequent self-occlusion. To address this gap, we introduce InfantNet, the largest real-image infant dataset to date, comprising 108,902 RGB images of infants aged 6-18 months. Each image is annotated with 2D keypoints, and a curated subset of 11,642 images additionally includes 3D pose and shape annotations with full SMIL parameters, verified through multi-view rendering and reprojection-error filtering. We use an iterative annotation pipeline to ensure high fidelity across both 2D and 3D labels. InfantNet establishes a large-scale, comprehensive benchmark for infant 2D keypoint detection and 3D pose-and-shape recovery. Baseline experiments demonstrate that state-of-the-art adult pose estimators do not generalize well to infants, whereas fine-tuning on InfantNet yields a consistent improvement. The gains are even more pronounced for 3D pose and shape estimation, where SMIL-based estimation demonstrates robust performance with accuracy comparable to results reported for SMPL-based models. By releasing the InfantNet dataset and benchmark, we provide a vital resource for advancing infant pose analysis and related healthcare applications.

cs.CV↗
arXiv2026

SkillCT: A Multidimensional Benchmark and Execution-Free Baseline for Agent Skill Assessment

Agent skills package instructions, scripts, and resources for specialized tasks and are increasingly distributed through public Web repositories. However, existing skill benchmarks primarily focus on downstream task execution, insufficiently assessing specification quality, behavioral robustness, semantic alignment, and responsibility boundaries. Their reliance on task-specific tools, environments, and repeated executions limits scalability and pre-deployment assessment. To address these challenges, we introduce SkillCT, a benchmark that broadens skill assessment from downstream utility to a multidimensional characterization of skill quality. SkillCT evaluates skills across nine complementary dimensions spanning three aspects: Static Assessment, which captures intrinsic properties of the skill specification; Dynamic Assessment, which measures whether the induced behavior remains stable across backbone models and request variations; and Human-Assisted Assessment, which examines behavior-chain consistency and task granularity. These dimensions provide a broader view of skill quality than execution success. Applying SkillCT to a public skill corpus establishes a reference benchmark for analyzing quality weaknesses and relationships across dimensions. Building on this benchmark, we propose an execution-free quality evaluation baseline that predicts the multidimensional quality profiles of unseen skills by retrieving and leveraging similar reference skills, without executing the skills under evaluation. This reduces token consumption and evaluation time, while preserving consistency with the full assessment. Furthermore, SkillCT-guided optimization improves downstream task success and execution efficiency, demonstrating its value for skill refinement. These findings highlight the effectiveness of SkillCT in supporting scalable skill assessment, optimization, and repository maintenance.

cs.AI↗
arXiv2026

Visually Irrelevant Hallucinations in Medical Vision-Language Models

Medical vision-language models (VLMs) show strong performance in radiology report generation (RRG) and medical visual question answering (Med-VQA), producing free-form clinical descriptions and answers from medical images. However, they may generate clinically meaningful statements that lack visual evidence or contradict image content. We refer to these errors as visually irrelevant hallucinations and investigate them at the phrase level. We propose CLEAR (Counterfactual Localized Evaluation of Alignment and Response), a model-agnostic post-hoc framework for phrase-level hallucination risk estimation. CLEAR decomposes generated responses into visually assessable clinical phrases. For each phrase, phrase-conditioned grounding identifies a candidate visual region, which is selectively perturbed by a localized counterfactual intervention. CLEAR characterizes the intervention using two complementary auditing cues: response cues, capturing semantic and clinically meaningful changes in the generated response, and alignment cues, measuring changes in image-phrase compatibility. These cues are normalized and jointly modeled to estimate phrase-level hallucination risk. Across four datasets spanning RRG and Med-VQA, CLEAR outperforms state-of-the-art hallucination detection baselines across multiple medical and general-purpose VLMs. On MIMIC-CXR, CLEAR improves AUPRC from 85.10% to 90.23%. Its effectiveness is further validated on a GT-derived controlled VIH benchmark. These results establish visually irrelevant hallucination as an important fine-grained reliability failure mode in medical VLMs and demonstrate the value of model-agnostic visual-support auditing for more trustworthy multimodal clinical generation.

cs.CV↗
arXiv2026

SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?

Agent skills, structured procedural knowledge packages injected at inference time, are increasingly used to augment LLM agents on software engineering tasks. However, their real utility in end-to-end development settings remains unclear, and no standard procedure tests an individual skill before installation. We present swe-skills-bench, a skill-centric, requirement-driven benchmark that measures how much individual agent skills improve performance on real-world software engineering (SWE) tasks. We start from 98 public SWE skills, each paired with a real GitHub repository pinned at a fixed commit. For each skill, we derive repository-specific requirements, create executable verifiers, and build a reproducible environment. Under each of four agent-model configurations, we run every task once without the skill and once with it, yielding 3,920 paired evaluations. For each requirement, we compute a paired delta, i.e., the change in outcome when the skill is added. We then aggregate these deltas across a skill's ten requirements into a per-skill utility profile, and use a statistical test to classify each skill as Golden (significantly helpful), Ineffective (no significant effect), or Harmful (significantly detrimental). Our results show that skills provide limited benefits: 92 of 98 skills have no statistically significant effect, and adding skills raises the average pass ratio only from 76.6% to 76.9%. The few helpful skills supply procedures that the agent would otherwise omit (gains of up to $+17.7$ points). In contrast, harmful skills anchor the agent on their own templates and override the task's explicit instructions (losses of up to -17.0 points). Our code is available at https://github.com/ControllableModel/SWE-Skills-Bench.

cs.SE↗
arXiv2026

Understanding Non-Split 2-Group Symmetry: (3+1)D SymTFT, Anomaly and Bordism

We present a systematic study of finite, non-split 2-group symmetries with a non-trivial Postnikov class, focusing on the simplest example with a $\mathbb{Z}_2$ 0-form symmetry and a $\mathbb{Z}_2$ 1-form symmetry, intertwined together by the non-trivial Postnikov class in $H^3(B\mathbb{Z}_2;\mathbb{Z}_2)\cong\mathbb{Z}_2$, denoted by $\mathcal{G}$. We classify the anomalies of this 2-group symmetry for physical theories in $d$-dimensional spacetime, by computing the oriented bordism groups $Ω_{d+1}^{\rm SO}(B\mathcal{G})$ and the spin bordism groups $Ω_{d+1}^{\rm Spin}(B\mathcal{G})$ for $d\leq 5$. For the case of $d=3$, we investigate the Symmetry TFT/TO of the 2-group symmetry $\mathcal{G}$ using the language of fusion 2-categories, as well as (3+1)D TQFT actions, for the cases without or with the 2-group anomaly classified by $\operatorname{Hom}\left(\widetildeΩ_4^{\rm SO}(B\mathcal{G}),U(1)\right) \cong H^4(B\mathcal{G};U(1))\cong\mathbb{Z}_2$. We classify the minimal topological and physical boundary conditions of the Symmetry TFTs, and carry out the categorical Landau paradigm for such a non-split finite 2-group.

hep-th↗
arXiv2026

Recall or Relearn? A Mixture of In-Context Experts for Recurring Concept Drift

Tabular foundation models (TFMs) are being applied to Web data streams such as network intrusion detection, where they adapt by updating the labelled rows of their context without retraining. When a past concept returns after leaving the context, a TFM faces a recall-or-relearn decision between restoring stored rows of the concept and relearning it from new labels. We propose MICE (Mixture of In-Context Experts), a training-free learner that cuts the stream into segments at detected changes. It merges a closed segment into a stored context when the TFM's estimated additional error on the segment is within a tolerance, under a capacity limit on stored contexts. A fast level weights the stored contexts and first-in-first-out (FIFO) windows by the log-loss of their predictions on the last labelled batch. We show that, with pure stored contexts and a margin condition, this fast level gains over relearning, in expected accuracy, the normalised positive area between the stored context's accuracy and the TFM's learning curve, up to a first-batch term. A slow level mixes the fast level's prediction with that of the FIFO window of the last labelled rows and bounds the discounted half-Brier regret of MICE against this window at every time. Learning curves measured beforehand predict the ordering of the observed recall gains. MICE outperforms reset, which empties the context at detected drifts, on grid streams with random boundary phases. Merging keeps fewer experts than pools that store every segment, and with fixed segmentation a time-matched snapshot pool matches MICE on the shifted grid. On real streams, MICE has the highest mean accuracy of all policies we compare.

cs.LG↗
arXiv2026

Hallucination-Aware Multi-Agent RAG via Internal Attention Rewards

Retrieval-augmented generation (RAG) grounds language models in retrieved evidence, but its outputs often contain claims unsupported by that evidence. Recent multi-agent RAG systems decompose the pipeline into cooperative decisions, yet they typically optimize all agents under a shared task reward such as EM or F1, which measures answer overlap rather than faithfulness to evidence. We propose a hallucination-aware multi-agent RAG framework that combines the task reward with an internal hallucination-risk signal computed from the generator's attention-head norms. Three role-specific LoRA adapters, a query planner, an evidence selector, and an answer generator, share a frozen backbone. A lightweight critic is trained once on TruthfulQA truthful/false pairs and then kept frozen, providing a complementary risk score that is added to the shared reward during cooperative PPO training. Across three backbones (Qwen3-8B, Llama3-8B, Llama2-13B) and three benchmarks (RAGTruth, HotpotQA, 2WikiMultiHopQA), our method improves task performance and reduces hallucination rates over the strongest multi-agent baseline. Ablations and critic diagnostics show that the TruthfulQA-calibrated signal provides useful reward shaping that transfers across RAG tasks.

cs.IR↗
arXiv2026

RIGen: Beyond Action Following with Reciprocal Interaction in Generative Gameplay

Generative gameplay requires more than immediate responses to player actions.Actions should have persistent consequences, while the generated world should act according to its own state and rules, shaping the player's situation and subsequent choices. Video generation captures local dynamics, but limited history, partially observed game state, and the need for contextual behavior decisions make such interactions difficult to sustain through audiovisual prediction alone. We introduce \rig (Reciprocally Interactive Generation), a framework for reciprocal interaction in generative gameplay. \rig combines persistent state and configurable rules, contextual interaction reasoning, and action-conditioned audiovisual generation. Together, these components make interaction consequences and actor behavior jointly determine generated content while maintaining consistency between game state and visible outcomes. We construct a Xonotic gameplay dataset to train an audiovisual generator for learning dynamics and interactions. Further post-training enables \rigarena, an interactive generative first-person shooter (FPS) prototype. Experiments demonstrate more reliable combat outcomes and reciprocal interaction between the player and the generated world. Project page: https://murphylmf.github.io/RIGen/.

cs.CV↗
arXiv2026

LAW: Unifying Language, Action, and World Modeling for Agentic Physical Intelligence

World-action models jointly predict actions and future world states, but long-horizon manipulation further requires language reasoning to plan subtask objectives, assess execution progress, and guide action selection. We present LAW, a closed-loop VLM--WAM framework that unifies language reasoning, action generation, and world modeling through semantic subtask progress. Within the LAW agent, a hybrid-expert vision-language model plans the remaining subtasks, estimates the values of and fuses actions proposed by a world-action model, and asynchronously assesses execution status to support continued execution and subtask transitions. Within the world-action model, Mixture-of-Transformer Experts integrates modality-specific and shared computation for joint video--action modeling. Dual-stream hierarchical visual memory maintains temporal context for both models, while subtask-local value learning aligns action evaluation with semantic objectives and preserves supervision from successful prefixes of failed trajectories. LAW achieves 94.21% average success across 50 RoboTwin tasks. Replacing the WAM backbone with OpenWAM-$alpha$ further raises the average success rate to 95.56%, establishing a new open-source state of the art on RoboTwin. Despite having only 2B parameters, our VLM also outperforms variants using Embodied-R1.5 and GPT-5.5 High. On three long-horizon real-robot tasks, LAW achieves 65% average success, compared with 28.33% for $pi_{0.5}$ and 25% for Fast-WAM.

cs.RO↗
arXiv2026

Dynamic Query as Budgeting for Tiny Object Detection

Recent DETR-based tiny object detectors adopt dynamic query mechanisms to handle severe density imbalance. However, existing designs entangle two effects: how many and which queries are decoded. This entanglement obscures what drives the gains and complicates dynamic-query design; additional learned predictors and heuristic filtering also introduce overhead and potential selection errors. We disentangle these effects and show that dynamic query primarily acts as \emph{budgeting}---allocating decoder query capacity according to object density---rather than intricate query selection.Motivated by this view, we formulate a monotonic and conservative budgeting principle and propose a budgeting-based DETR (BUTR), a streamlined framework that uses a lightweight, non-learned budgeter to estimate an input-dependent query budget $K(x)$ from encoder confidences. BUTR retains standard Top-$K(x)$ query initialization without additional learned modules or heuristic filtering in the budgeter.At $800\times800$ resolution, BUTR-L achieves 35.7 AP at 47.8 FPS on AI-TOD-v2 and 29.8 AP at 42.1 FPS on VisDrone on an NVIDIA T4 GPU. Inference-time integration on COCO further demonstrates its applicability to general scenario detection. Code is available at https://github.com/zhangyi-magic/BUTR.

cs.CV↗
arXiv2026

In-Orbit Computing for Remote Sensing: Task-Oriented Processing and Transmission

The growing volume and timeliness requirements of remote sensing data are exposing the limitations of the conventional "transmit-then-process" architecture, particularly under intermittent satellite-ground connectivity, constrained downlink bandwidth, and limited onboard computing resources. This article presents a semantic-aware space-ground collaborative remote sensing (SSGC-RS) framework, in which remote sensing data are processed to different semantic depths according to task requirements and available communication-computing resources. Three operating modes are considered, ranging from fully onboard task inference to semantic representation delivery and space-ground split inference. A land-cover recognition case study illustrates how these modes address different communication and computing bottlenecks. Finally, emerging challenges and future directions for intelligent remote sensing systems are discussed.

eess.SP↗
arXiv2026

QwenGyre: An Elastic Reinforcement Learning Framework for Training xLong-Horizon Agents

Large language model (LLM) agents increasingly undertake extreme-long (xlong) horizon tasks, where a single execution can span hours, hundreds of model--environment interactions, and nearly 1M tokens per rollout. Applying online reinforcement learning (RL) to such executions poses two fundamental challenges: (1) severe execution variance and prolonged rollout delays cause massive GPU idling; and (2) complex non-linear branching generates massive trajectory redundancy, crippling training efficiency. To address these, we presents QwenGyre, an end-to-end framework for xlong-horizon online RL. QwenGyre elastically reallocates GPUs between rollout and training without interrupting live executions, while its trajectory processor reconstructs branching histories, scores partial progress, and deduplicates redundant paths to bound training costs. Scaled to our flagship model, Qwen~3.8 2.4T, with 700K tokens per rollout, QwenGyre yields a 6.0% absolute gain on NL2RepoBench (52.5% $\to$ 58.5%) in 48 steps. Across our evaluations on diverse domains of training datasets, QwenGyre delivers up to $1.85\times$ and $1.78\times$ speedups over Colocate and Async, respectively.

cs.LG↗
arXiv2026

ACE-Compression: An Ability-Centric Evaluation of Compressed Large Language Models

Model compression makes large language models deployable, but today's evaluation tells an incomplete story. Standard practice measures retained task performance and stops there. This misses a critical fact: compression perturbs the distributed knowledge embedded across billions of weights, and task scores alone cannot reveal which underlying abilities survive and which quietly collapse. To solve this question, we introduce ACE-Compression, an ability-centric evaluation framework. Instead of asking only "does the compressed model still answer correctly?", we ask "which of its core abilities remain intact?" Our framework probes four deployment-facing dimensions: procedural reasoning and constraint execution, knowledge grounding and truthfulness, social cognition and affective understanding, and persona fidelity and stylistic consistency. Across more than 35 compressed checkpoints from six open-weight LLMs-covering quantization and pruning-we find that compression redistributes ability profiles. Coarse, local, and surface-level signals often persist, while fine-grained, long-tail, and context-dependent abilities become fragile. Quantization consistently outperforms pruning in retention, but the practical meaning of that gap depends entirely on which ability an application requires. Based on our findings, deployers can choose compression methods by the ability profile their application needs, not just by a leaderboard average. And algorithm designers gain diagnostic signals to build compression methods that preserve what matters. We believe that ACE-Compression turns compression from a black-box accuracy drop into an interpretable intervention-one we can measure, understand, and ultimately improve.

cs.AI↗
arXiv2026

Second-order difference-frequency effects on floating wind turbine responses: full-QTF modelling and experimental comparison

Accurate prediction of low-frequency responses is important for the dynamic assessment of floating offshore wind turbines (FOWTs), as second-order difference-frequency wave loads can contribute substantially to platform motions and mooring loads. Their relative importance, however, remains less well established when aerodynamic and hydrodynamic low-frequency excitations coexist under operating conditions. This study develops an efficient time-domain formulation for full difference-frequency quadratic transfer function (QTF) loading using bandwidth truncation and integrates it into a fully coupled model of the IEA 15 MW turbine on the VolturnUS-S semi-submersible platform. Numerical predictions are compared with 1:70-scale real-time hybrid wave-basin experiments covering both irregular wave-only and combined turbulent wind-wave conditions. Three hydrodynamic formulations are compared: linear wave excitation, Newman's approximation, and the full-QTF model. Under wave-only conditions, second-order difference-frequency loading substantially improves the prediction of surge and the associated mooring response, particularly in energetic sea states, with the full-QTF model generally providing the closest agreement with measured low-frequency spectra and peak-amplitude distributions. Under combined wind-wave conditions, low-frequency thrust fluctuations dominate near rated wind speed, reducing the relative importance of difference-frequency wave excitation, whereas its contribution becomes more apparent at higher wind speeds as thrust fluctuations diminish. Overall, the full-QTF model provides the most consistent improvement over the linear and Newman models, with maximum surge standard deviation discrepancy from measurement under the most energetic wave conditions less than 18%, with smaller discrepancies for heave, pitch, and mooring tensions.

physics.flu-dyn↗
arXiv2026

AgentPolicyForge: From Executable Environments to Policy Benchmarks and Adaptive Guards

Customized agent deployments define policies over tools, resources, and execution conditions. Providing these policies in the task agent's prompt does not reliably prevent violations. Training a guard requires examples of tool calls that comply with or violate those policies. We introduce AgentPolicyForge, a framework that uses executable environments to construct policy benchmarks and iteratively train policy guards. With this framework, we construct PolicyGym, a dataset across ten policy families using 698 scenarios for training data generation and 254 separate held-out scenarios for human-reviewed test sets. Tool calls are executed and their outcomes recorded to help determine whether they comply with the policy. Every violating case uses a permitted tool, requiring checks of arguments, resources, and context rather than tool identity alone. The case construction process supports iterative guard training: an attacker generates requests using retrieved strategies, and the guard is fine-tuned on balanced, environment-verified examples. Training improves policy enforcement even with a fixed archive. Updating the archive further reduces in-domain attack success compared with keeping it fixed. We compare policy-aware task agent self-regulation with a trained policy guard. On PolicyGym, the final guard-based configuration achieves a 40-44% relative reduction in attack success rate and fewer benign rejections than policy prompting alone.

cs.AI↗
arXiv2026

Safe, or Simply Incapable? Rethinking Safety Evaluation for Phone-Use Agents

When a phone-use agent avoids harm, does that show safety, or simply inability to act? Existing evaluations often cannot tell. A harmful outcome may be avoided because the agent recognized the risk and chose the safe action, or because it failed to understand the screen or execute any relevant action at all. These cases have different causes and call for different fixes, yet current benchmarks often merge them under task success, refusal, or final harmful outcome. We address this problem with PhoneSafety, a benchmark of 700 safety-critical moments drawn from real phone interactions across more than 130 apps. Each instance isolates the next decision at a risky moment and asks a simple question: does the model take the safe action, take the unsafe action, or fail to do anything useful? We evaluate eight representative phone-use agents under this framework. Our results reveal two main patterns. First, stronger general phone-use ability does not reliably imply safer choices at risky moments. Models that perform better on ordinary app tasks are not always the ones that behave more safely when the next action matters. Second, failures to do anything useful behave like a capability signal rather than a safety signal: they track broader phone-use performance and increase when visual input is removed. Across models, failures split into two recurring patterns: unsafe choices in settings where the model can act but chooses wrongly, and inability to act in more visually and operationally demanding screens. Overall, a harmless outcome is not enough to count as evidence of safety. Evaluating phone-use agents requires separating unsafe judgment from inability to act. Code and public data: \url{https://github.com/tangzhy/PhoneSafety}.

cs.CL↗
arXiv2026

Leaky-Wave Antenna for Non-Invasive Continuous Glucose Monitoring

Non-invasive continuous glucose monitoring requires sensing techniques capable of delivering stable, accurate, and real-time quantification of glucose-induced dielectric variations. This paper presents a microwave-based methodology that employs a leaky-wave antenna (LWA) to launch localized traveling electromagnetic waves and estimate glucose concentration through phase measurements at a fixed operating frequency. The LWA architecture enhances electromagnetic coupling with the material under test through controlled energy leakage, enabling strong linearity and high phase sensitivity within the physiological permittivity range (relative permittivity 68-70). In vitro measurements using fetal bovine serum-glucose samples demonstrate a total phase shift of 8.51 degree across glucose concentrations from 50 to 255 mg per liter. Preliminary in vivo experiments conducted on two human subjects show strong correlations between measured S11 phase and reference glucose levels (r=0.939 and r=0.906), confirming the potential of the LWA-based sensing mechanism to track dynamic glycemic variations in a proof-of-concept setting. These results highlight the potential of the proposed approach for practical, continuous, and non-invasive glucose monitoring.

physics.app-ph↗