Search arXiv⌕ Search

arXiv subjects

Zijian Wang

Publications and source records attributed to Zijian Wang.

At least 19 recordsLinked to original sources
arXiv2026

When Charts Deceive Web Agents: A Paired Benchmark for Misleading Visualization-Grounded Tasks

Web agents increasingly browse, interpret, and act on web content on behalf of users. In many realistic workflows, however, the visual evidence they consume is not neutral: charts and dashboards can be misleading, shaping judgments without issuing any explicit malicious instruction. Unlike chart question answering, Web Agents treat charts as intermediate evidence whose misinterpretation can propagate into concrete downstream actions --- a reliability risk that remains underexplored. We introduce MisVisAgentBench, the first paired executable benchmark for evaluating how misleading visualizations affect Web Agent execution. It comprises 280 task instances in matched clean--misleading pairs that differ only in chart evidence. We further provide a diagnostic framework that isolates misleading-induced failures from baseline task difficulty and captures execution-level effects such as extra browsing steps and decision hesitation. Across thirteen vision-language Web Agents, replacing clean charts with misleading counterparts reduces the average task success rate by more than 34 percentage points. Moreover, these failures predominantly follow the targeted misleading action branch, suggesting that deceptive charts systematically redirect rather than randomly disrupt agent decisions. These results reveal that visualization robustness is a critical yet understudied dimension of Web Agent safety and reliability.

cs.CL↗
arXiv2026

Scattering-force-driven light-sheet fluorescence microscopy for scan-free three-dimensional imaging

Optical manipulation offers a powerful non-contact approach to precise positioning, directional transport and three-dimensional (3D) imaging of microscale objects. However, positional fluctuations during optical confinement introduce spatial sampling errors that compromise imaging quality and acquisition speed. Here, we demonstrate a scattering-force-driven optical manipulation imaging method that replaces the stepwise scanning of conventional optical trapping with continuous directional transport of microparticles. An axial scattering force generated by a low-numerical-aperture focused beam propels microparticles through a light sheet, while their positions and fluorescence signals are simultaneously recorded to establish direct spatial correspondence between particle motion and tomographic images. Experiments demonstrate that this approach reduces the effects of positional fluctuations on 3D imaging, achieving an axial sampling interval of 0.1 μm and an imaging rate of 250 Hz. The method also enables effective imaging of samples with diverse structural features while reducing optical power requirements by 70%. By combining robust spatial sampling, high-speed acquisition and reduced photodamage, this approach offers a promising route to rapid 3D imaging, dynamic observation and microscale spatial analysis of microparticles and biological specimens.

physics.optics↗
arXiv2026

IntentDelta: A Human-Constructed Benchmark for Evolving User Intent

Language-model agents must adapt as users revise their requirements and priorities during complex tasks. We introduce IntentDelta, a human-constructed multi-turn benchmark with 638 trajectories and 3,766 turns across WebShop and TravelPlanner. Each turn pairs a user utterance with a cumulative intention state and human reference actions. The trajectories are cross-verified and cover feasibility- and preference-driven changes in constraints and priorities. We evaluate eleven agents on intention recovery and action quality. Average Intention Recovery is 33.1\% in TravelPlanner and 44.7\% in WebShop, while average Action Score is 31.6% and 41.0%, respectively. Agents struggle to preserve earlier requirements and distinguish soft preferences from must-have constraints. Correct intention recovery also does not guarantee successful execution, highlighting challenges in translating evolving user intentions into actions.

cs.CL↗
arXiv2026

PosterHarness: Turning Scientific Poster Generation into an Auditable Instruction-Following Benchmark

Text-rich image models can now design poster-scale layouts, but we lack ways to measure whether they honor scientific communication contracts: legible labels, prescribed aspect ratios, and -- above all -- abstaining from fabricated scientific figures. We present POSTERHARNESS, an auditable harness reframing poster generation as measurable instruction-following tasks, with a pilot benchmark and failure taxonomy. POSTERHARNESS uses a placeholder-first contract to separate two jobs models otherwise conflate. The model performs visual-summary design: typography, reading path, color, and background -- but never draws data-bearing figures. Every figure region must be an empty labeled placeholder; a deterministic compositor inserts real source-paper figures at detected coordinates. This makes properties measurable: placeholder count and ID accuracy, blankness, aspect-ratio compliance, abstention from synthesized graphics, public-text hygiene, and source-figure provenance -- with failures logged as explicit rejections, not hidden in plausible-looking output. We instantiate the harness on 12 papers (6 HEP, 6 AI/ML-adjacent) and report three findings. (i) A counterfactual probe shows the placeholder contract drives VLM-counted synthesized figures from 34 to 0 across three papers. (ii) A failure taxonomy identifies blocking contracts: placeholder geometry, placeholder QA, template critic, and public text. (iii) Comparison with Paper2Poster shows a trade-off: PosterHarness yields higher-resolution artifacts, lower white-canvas fraction, and stronger VLM visual preference; the deterministic baseline retains slightly more PosterQuiz-style information and runs faster. We report this as regime characterization, not a superiority claim. All artifacts, prompts, manifests, and audit scripts are released as a reusable evaluation component.

cs.CV↗
arXiv2026

PulseBound: Future-Beat State Forecasting Under an Explicit Information Boundary

Predictive representation learning from photoplethysmography (PPG) can violate causal information access even with causal attention, as normalization, nonlocal transforms, or companion views may depend on withheld samples. We introduce PulseBound, a PPG representation learner combining physiologically structured future-beat prediction with an explicit stored-window information boundary. A content-independent cutoff separates the visible prefix from the prediction target. Prefix-only normalization, suffix replacement before derived-view construction, and aligned masking ensure that encoder inputs depend only on the visible prefix and cutoff. This yields stored-suffix invariance: with fixed model state, randomness, prefix, and cutoff, changing the stored suffix cannot change the forecast context. A shared horizon-conditioned head predicts nine rhythm and morphology descriptors for up to four extractor-valid future beats, using elementwise validity masks; optional ECG-derived pulse-arrival-time supervision is restricted to training. On MIMIC and VitalDB groups held out from PulseBound backbone pretraining, PulseBound reduces nine-state transformed-space MAE relative to last-visible-beat persistence by 28.06% and 22.22%, respectively, with gains in MAE, MAE-Skill, and Spearman correlation across all 40 source-cutoff-horizon cells. In a separate comparison of seven models on 13 downstream tasks, PulseBound achieves the best mean on nine frozen linear-probe and seven full-fine-tuning tasks. Stored-suffix interventions cause zero recorded changes in forecast contexts or predictions, with zero suffix-input gradients at audited precision under the stored-window interface. These findings separate three testable aspects of predictive physiological representation learning: information access, supervised future structure, and transfer.

cs.AI↗
arXiv2026

From Earth to Mars and Beyond: AI-Driven Cosmic-Ray Validation and Prediction with CorsikaAgent

We calculate the sea-level cosmic-ray muon flux with CORSIKA and develop CorsikaAgent, an auditable large-language-model-based system for configuring, executing, and validating air-shower simulations, and for predicting particle fluxes in unexplored environments. The physics calculation uses proton primaries described by the solar-minimum spectrum of Papini et al., with UrQMD at low interaction energies and either Sibyll 2.3e or QGSJET-III at high energies. The predicted momentum spectra at approximately 0 and 75 degrees are compared with experimental measurements of the atmospheric muon flux and with commonly used analytic parametrizations. The two interaction-model configurations give similar spectra and describe the principal momentum and zenith-angle dependences of the data. The agent independently reconstructs the same observable through a different sampling strategy and keeps every steering card, executable, seed, log, and analysis script bound to the result it produced; its output is consistent with the direct calculation at the level resolved by the present comparison. Transferred to Mars, the workflow reproduces the CORSIKA 8 benchmark, and transferred to Venus it produces the first prediction of the surface muon flux, an environment for which no measurement exists. The agent thus extends the workflow from auditable reproduction to first-principles prediction: it autonomously reconfigures the simulation(atmosphere, magnetic field, electric field)for new environments, while the proton-only primary model, limited high-momentum statistics, and an incomplete systematic-uncertainty treatment bound the physical precision that can currently be claimed.

hep-ph↗
arXiv2026

Quasilinear Equations with Exponential Reactions and Measure Data on Carnot Groups

In this paper, we establish existence and pointwise Wolff potential bounds for nonnegative solutions of $-Δ_{\mathbb{G},p}u=H_l(αu^β)+μ$ on Carnot groups, where $H_l(t)=e^t-\sum_{j=0}^{l-1}t^j/j!$ and $μ$ is a nonnegative Radon measure. Under suitable conditions on the reaction parameters, we treat arbitrary open sets in the subcritical range $1<p<Q$ under a maximal potential condition, and bounded domains in the critical case $p=Q$ under a small total mass condition.

math.AP↗
arXiv2026

ShelfChange3D: Object-Level 3D Change Detection for Retail Shelf Monitoring

Reliable shelf monitoring is an important capability for retail automation, yet existing out-of-stock detection methods mainly operate in image space and lack metric 3D localization for downstream robotic systems. We formulate shelf monitoring as object-level 3D change detection: given two RGB-D observations captured at different times, the goal is to identify changed products and localize each change with a 3D bounding box. To support this task, we introduce ShelfChange3D, comprising 145K synthetic and 5K real-world paired RGB-D observations with object-level 3D change annotations. We further propose ChangeBox, an end-to-end framework that jointly reasons over paired observations and predicts object-level 3D change boxes. To improve localization accuracy, we introduce a geometry-based refinement stage that exploits depth and gravity prior to estimate relative pose and refine predicted boxes. Experiments show that ChangeBox outperforms existing change detection baselines, with further gains from refinement and effective transfer from synthetic to real-world observations.

cs.CV↗
arXiv2026

FedSPM: Routing-Enabled Federated Learning under Dual Heterogeneity via Semiparametric Mixture

Routing-prediction federated learning has emerged as a new paradigm that reframes inter-client heterogeneity as a resource for system-level intelligence: at inference time, the server routes each external query to the best-matched client for prediction. Existing approaches, however, typically treat each client as internally homogeneous, overlooking latent subpopulations within local data. For example, patients with the same diagnosis at one hospital may exhibit morphologically distinct disease subtypes. The coexistence of inter-client and intra-client heterogeneity, which we call dual heterogeneity, can impair both routing and prediction. To address this challenge, we propose FedSPM, a routing-enabled semiparametric mixture framework that represents each client using client-specific latent components. Each component combines a predictive distribution for classification with a feature distribution for routing. To flexibly model feature distributions while effectively sharing information across clients, FedSPM models their density ratios relative to a common nonparametric measure estimated via empirical likelihood. We develop a federated expectation-maximization algorithm that optimizes a tractable surrogate and prove convergence of the exact profiled objective at the standard $\mathcal{O}(1/\sqrt{T})$ rate when the surrogate errors are properly controlled. Experiments on controlled benchmarks and real-world medical data demonstrate consistent improvements in routing and prediction under dual heterogeneity. Code is available at https://github.com/zijianwang0510/FedSPM.

cs.LG↗
arXiv2026

DeepSurvey: Agent-Oriented Automated Survey Generation with Analytical Depth and Citation Reliability

As scientific literature grows rapidly and research increasingly involves AI agents, automated survey generation has become a key capability for both agents and human researchers. For such agents, a survey serves as a primary knowledge source of a field prior to research. Since pretrained models already encode broad knowledge of established work, these consumers benefit more from analytical depth than from breadth alone; moreover, unsupported claims, once ingested as knowledge, can propagate into downstream research. However, existing systems tend to overemphasize coverage and presentation; they suffer from limited analytical depth due to reliance on abstracts and isolated paper processing, and from unreliable citations due to imprecise retrieval and post-hoc grounding. We present DeepSurvey, an agentic generation system that addresses both limitations. To enhance depth, DeepSurvey extracts structured keynotes, models cross-paper relationships through clustering and comparative analysis, and integrates a code-agent subsystem to recover implementation-level details. To fortify reliability, it combines citation-graph expansion with hybrid filtering for topic-focused retrieval, enforces evidence-constrained analysis and writing, and deploys multi-granularity agentic refinement to validate citation--claim alignment. Experiments show that DeepSurvey achieves the highest content score (8.34/10) and citation quality (recall and precision gains of 25.3\% and 35.2\% over the strongest baseline), generalizes more robustly across domains, and is preferred by domain experts over human-written surveys (83.3\% in overall quality, 100\% in content depth). Moreover, when a coding agent uses a survey as its only literature source, the agent equipped with DeepSurvey achieves the best performance among human-written and baseline-generated surveys.

cs.AI↗
arXiv2026

Production of lepton-flavor-violating scalars through resonant positive-muon annihilation on atomic electrons

We investigate an invisible lepton-flavor-violating scalar $ϕ$ with exclusive $e-μ$ couplings and study its resonant production via $μ^+e^-\toϕ$ in fixed-target experiments. Since the effective center-of-mass energy is determined by the momentum of the initial-state bound electrons, atomic effects can significantly affect the resonance behavior. We therefore employ relativistic bound-state electron wave functions to calculate the production cross section and reveal a material-dependent broadening of the resonance lineshape. For the proposed HIAF experiment, fewer than one day of data taking ($6\times10^{10}$ MOT) can probe couplings at the $10^{-5}$ level at 90\% confidence level near resonance, demonstrating that high-intensity muon fixed-target experiments provide a powerful complementary probe of lepton-flavor violation.

hep-ph↗
arXiv2026

Tracing the Evidence: Faithful Token Attribution Through Vision-Language Reasoning

Large vision-language models (LVLMs) exhibit strong reasoning capabilities, yet the visual and textual evidence supporting the generated responses remains difficult to identify. Faithful token attribution explains an LVLM's response by assigning scores that rank image and prompt tokens by how much the model relies on them, such that removing higher-ranked tokens causes the likelihood of the generated response to drop more rapidly. However, existing token-attribution methods have been developed mainly for text-based language models, and our empirical study reveals two challenges when complex multimodal sources are involved. First, the joint image-text attribution can underrepresent visual evidence relative to text, obscuring the image regions supporting the response. Second, visual evidence may influence the generated response through multiple intermediate reasoning paths, while existing methods trace only a limited subset of these paths, causing important visual contributions to be underestimated. Motivated by these insights, we introduce VTrace, a multimodal token-attribution framework that traces input contributions through intermediate reasoning and calibrates attribution scores across modalities. VTrace constructs pairwise attributions that highlight token-specific contributions and aggregates all forward attribution paths in closed form to account for both direct and indirect contributions. Cross-modal calibration then rescales image and text attribution scores using modality contributions estimated from response-likelihood changes, enabling a unified ranking of input tokens. Evaluations against seven baselines across six visual reasoning benchmarks demonstrate the superior attribution faithfulness. Project page: https://vtrace-attribution.github.io/.

cs.CV↗
arXiv2026

E2E-SWE: Benchmarking LLMs on Building Working Codebases from Scratch

Coding agents powered by large language models (LLMs) are evolving from making localized code changes to developing complete software repositories. However, evaluating repository-scale generation remains challenging: tasks must demand system-level reasoning while ensuring that all evaluated behaviors are precisely specified and independent of any particular implementation. We introduce E2E-SWE, a benchmark for evaluating whether coding agents can build complete, functional software repositories end to end. E2E-SWE contains 186 whole-repository generation tasks spanning 11 programming languages. Given only a natural-language specification and an empty workspace, an agent must implement a complete, installable project that satisfies a comprehensive suite of hidden tests. Each task is constructed by a software engineer in collaboration with an LLM; together, they develop the test suite and a corresponding implementation-independent specification. To ensure that tasks are well specified and practically solvable, we further subject them to an iterative verification process in which autonomous agents audit and repair task defects using static inspection and failures observed from real model rollouts. Evaluating 13 frontier models, we find substantial variation in end-to-end repository generation ability, with pass@1 ranging from 11.7% to 67.7%, providing strong model differentiation while leaving considerable headroom for future progress. Analysis of agent trajectories further reveals long, front-loaded reasoning patterns, highlighting the planning and system-level reasoning required to construct working codebases from scratch.

cs.SE↗
arXiv2026

Agentic Hybrid RAG for Evidence-Grounded Muon Collider Analysis

Muon collider research spans accelerator physics, detector instrumentation, and high-energy phenomenology, with relevant evidence scattered across a rapidly expanding and heterogeneous body of scientific literature. As high-energy physics (HEP) increasingly explores agent-assisted analysis workflows, efficiently locating, integrating, and verifying scientific evidence becomes an essential capability. While retrieval-augmented generation (RAG) offers a promising framework for scientific question answering, integrating agentic reasoning without compromising retrieval precision remains a key challenge. In this work, we present agentic hybrid RAG, an evidence-grounded RAG framework for muon collider research. The framework combines a hybrid retriever, integrating sparse lexical and dense semantic retrieval, with an agentic reasoning module for query decomposition, evidence expansion, and grounded answer generation. To enable systematic evaluation, we construct the first benchmark for retrieval-augmented scientific question answering in the muon collider domain, comprising a curated literature corpus together with dedicated retrieval and answer-generation benchmarks covering major detector and physics research topics. Extensive evaluation shows that hybrid retrieval provides the strongest retrieval backbone, while agentic reasoning is most effective for controlled evidence expansion and answer synthesis. Built on this principle, agentic hybrid RAG consistently outperforms representative retrieval and RAG baselines in retrieval effectiveness, answer quality, evidence coverage, and factual grounding. Together, the benchmark and framework provide a foundation for evidence-grounded scientific question answering and future HEP analysis agents operating over large-scale scientific literature. Code is available at \href{https://github.com/AItutorialjrb/RAG_muon_JINST}{this URL}.

hep-ex↗
arXiv2026

Making AI Scientists Auditable from Evidence to Claim

AI research systems can report improved results without establishing that the tested implementation matches the method described in their conclusions. We introduce Xcientist, a research harness that connects literature review, component-based idea generation, experiment validation and reporting. An evidence graph retains source passages, methods, baselines and evaluation conditions; component records track design revisions; and validation contracts specify the implementation, comparisons and artifacts required at each stage. We define claim drift as an unresolved mismatch between an accepted claim and its attributed sources, code or experiments. A retrospective audit across agent memory, traffic forecasting and physics-informed learning found audit claim drift rates of 3.6--16.7\% for Xcientist, lower than those of each of three comparison systems in all nine task-by-relation comparisons. This measure includes both confirmed mismatches and unverifiable claims; most of the difference arose from fewer unverifiable records. The case studies also document performance gains, unsuccessful revisions and limits on component attribution. This technical report describes the system, its implementation and the records needed to check how a proposed idea became a tested method and a final claim.

cs.AI↗
arXiv2026

Train Together or Merge Later? Unifying VLA Experts via a Shared Action Interface

Co-training offers a straightforward way to build a multi-task vision-language-action (VLA) policy, but can fall short of the performance achieved by training each task independently. The challenge is to retain these task-specific gains in a multi-task policy without joint post-training. Combining independently trained experts through model merging is a natural approach, yet strong individual experts do not necessarily yield a strong merged policy. We identify one source of this incompatibility: task-specific changes to the action interface, comprising action normalization and the action encoder and decoder. We propose PolicyWeave, combining merge-compatible post-training with context-guided sparse merging. During post-training, all experts retain the common base policy's action interface, while task adaptation is restricted to LoRA updates in the hidden layers of the action model. This makes the experts more compatible with existing model merging methods. However, merging all experts can still introduce interference from unrelated tasks at deployment. PolicyWeave scores each expert's LoRA updates using the initial visual-language context, determines the expert set through leave-one-layer-out ranking stability, and forms a sparse weighted merge of the selected updates that remains fixed for current task. We evaluate PolicyWeave with GR00T N1.5 on 18 RoboCasa365 tasks, using only 10% of the target-task demonstrations for supervised fine-tuning (SFT). Preserving the shared action interface raises the average success rate across four static merging methods from 17.0% to 52.8%. PolicyWeave achieves 64.7% success with these SFT experts and 74.1% after task-specific reinforcement learning (RL), compared with 60.7% for joint RL. Further evaluations on LIBERO-10 and an AgileX Piper arm support the deployment of independently learned skills in long-horizon and real-world manipulation.

cs.CV↗
arXiv2026

RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents

Computer-use agents (CUAs) have advanced along two separate lines: graphical interaction and software development through code and the command line. Real digital work requires both, interleaved rather than stacked end to end. We study hybrid CUAs that autonomously decide when to explore an interface, implement software, and run and visually verify their artifacts. We introduce RecreationWorld, a five-platform framework built around recreation: given a running reference, an agent must discover its behavior and build a faithful implementation with no prescribed workflow. RecreationWorld provides reproducible environments on Ubuntu, macOS, Windows, Android, and Web, plus a unified harness with native GUI control and coding tools. The running reference serves as an oracle for hidden behavioral tests, providing execution-grounded rewards. We scale trajectory generation with high-quality open-source applications. Models trained on these trajectories improve across five out-of-distribution coding and hybrid computer-use benchmarks and more frequently verify their rendered outputs, providing evidence of transfer beyond recreation. For held-out evaluation, we introduce RecreationBench, comprising 250 diverse tasks across domains and platforms. Reference-grounded programmatic and visual assertions cover action-conditioned outcomes at multiple interaction depths; each is validated on the reference and by human reviewers before the suite is frozen for automatic scoring. GPT-6 Astra leads at 58.1% overall, but passes all programmatic tests on just 2.8% of tasks. Agents reproduce static interface structure more reliably than interactions and computed outputs, while generated applications remain smaller and more monolithic than their references. We release the benchmark, environments, and test suites.

cs.CL↗
arXiv2026

Towards stratified sampling for redistricting plans

Rapid algorithmic developments have accelerated the sampling of redistricting ensembles (balanced graph partitions), yet evaluating rare events and sampling complex target measures remains a core challenge due to the high-dimensional and combinatorial nature of the phase space. We address a prerequisite for stratified sampling on this space: constructing and diagnosing candidate strata with suitable coverage and overlap. We build a grammar on observed plans by clustering districts into representative ``letters'' and using them to form plan-level ``words.'' A partition of unity over these words gives a soft assignment of plans to strata and allows us to estimate stratum masses and an overlap-induced flux matrix. We demonstrate this computational pipeline using real-world congressional redistricting data from Connecticut and examine how strata learned from one target distribution behave under related distributions. The resulting construction provides a foundation for future stratified sampling on spaces of redistricting plans or balanced graph partitions. We do not implement a complete stratified sampler here; evaluating whether the proposed strata improve sampling efficiency or reduce estimator variance is left for future work.

physics.soc-ph↗