Searcharxiv⌕ Search

arXiv subjects

Wenyi Li

Publications and source records attributed to Wenyi Li.

At least 19 recordsLinked to original sources
arXiv2026

RFW-A: Agentic Procedural Generation of 400K Articulated Assets over 500 Categories

High-quality articulated assets are essential for embodied perception, simulation, and manipulation, yet constructing them at scale remains difficult. Manual annotation and procedural modeling require substantial human effort. Recent agentic systems automate asset construction through program synthesis, but repeat costly reasoning and refinement for every new asset. Our key observation is that assets within a category can share construction logic despite variations in structure and appearance. Agentic synthesis and procedural generation are therefore complementary: agents can construct reusable programs, while procedural execution reuses the resulting logic across many samples. We therefore present RFW-A, which combines the two in a new articulated-asset generation paradigm. Starting from existing articulated programs, RFW-A organizes reusable components and their dependencies into category-level generators. Agents test and refine these generators across sampled configurations and motion states. Once constructed and validated, the generators produce diverse assets through procedural execution with no further model calls or token charges. The estimated API cost of generator authoring averages $10.62, amortizing to $0.05 per asset over 200 assets, versus $1.24 for the sampled per-asset agentic synthesis baseline. Using RFW-A, we built RFW-A-400K, containing over 400K sim-ready articulated assets across more than 500 categories. Experiments show high structural and kinematic validity and stable execution across physics simulators. Beyond asset validity, training with RFW-A-400K improves part-level articulation estimation on seen and unseen categories, while policies trained in simulation with retrieved and program-edited assets transfer zero-shot to real articulated objects.

cs.CV↗
arXiv2026

Quant-Arena: Benchmarking LLM Agents for Quantitative Trading Research

Quantitative research is an iterative loop: a researcher proposes a market hypothesis, implements it as an executable signal, tests it on historical data, and decides what to keep. Existing benchmarks score formulas, code, or individual trading decisions, leaving open whether large language models (LLMs) can complete this loop and produce strategies that work beyond the research period. We introduce Quant-Arena, a benchmark and public leaderboard for end-to-end quantitative research on real A-share market data. A model constructs features and stock-scoring code, receives feedback only on 2022--2023, and freezes one strategy for assessment on the sealed year 2024. A cash-account evaluator applies trading constraints and costs, including market impact estimated from a simulated order book. Scores are dominated by realized excess return and complemented by signal quality and the diversity of features used by the strategy. We evaluate nine frontier models, among which GPT-6 Astra scores highest; yet no model beats the equal-weight market in 2024. Quant-Arena is open to submissions from any model or agent harness.

cs.CL↗
arXiv2026

EmbodiedEvo: Self-Evolving Coding Agent For Physical Intelligence

Coding agents can construct executable programs and revise them through feedback, offering a foundation for embodied control without target-task fine-tuning. Through a programmatic robot interface, the agent observes the scene, executes control programs, and adapts them to the resulting physical state, serving as the closed-loop policy. However, this within-episode adaptation does not automatically preserve experience across scenes. Our trajectory analysis shows that agents spend interaction and inference budgets establishing effective operations and identifying their preconditions, while fresh agents cannot reuse these discoveries. This motivates EmbodiedEvo, a self-evolving coding agent that distills physical feedback into reusable code Skills and diagnostic Lessons. Experience is acquired on one Playground scene per task, then frozen for independent agents in unseen scenes, without updating model weights. On RoboDojo-Sim, EmbodiedEvo achieves an average score of 44.48 and a success rate of 37.82\%, surpassing all officially verified results. Paired tests show the largest gains on memory-dependent tasks and successful reuse of coordinated procedures, while fine contact control remains difficult. Experience from stronger models also improves weaker executors: Astra's experience raises Sol and Luna's success rates by 8.86 and 5.50 percentage points over no experience. Compared with raw interaction history, Skills and Lessons achieve comparable task performance at 53.0\% lower execution cost.

cs.RO↗
arXiv2026

World Models' Last Exam in Physics

Video world models can produce visually convincing yet physically inconsistent sequences, raising concerns about their reliability for prediction and planning in embodied AI systems. Existing evaluations often rely on model-based judgments or reference videos, while direct physical tests largely focus on mechanics. We introduce World Models' Last Exam in Physics, a measurement-based benchmark for evaluating physical consistency in video world models. The benchmark comprises 40 controlled tasks spanning mechanics, optics, fluids, thermal and phase-change phenomena, electromagnetism, and surface tension. Each task pairs an initial image and a generation prompt with predefined physical criteria, enabling interpretable tests of observable physical relationships without requiring reference videos. Its evaluator combines task-observability screening with task-specific quantitative physical measurements. Experiments on eight video generation models across 1,280 videos reveal persistent physical inconsistencies and substantial variation across tasks, with the best model achieving an overall score of 57.76 out of 100. Evaluation on synthetic videos with known physical relationships provides evidence for the validity of the measurement module under controlled conditions. The evaluator also achieves higher agreement with human judgments than a direct vision-language model baseline in both within-task rankings and pairwise comparisons. By combining coverage across physical domains with scores grounded in measurable evidence and explicit measurement limitations, the benchmark provides an interpretable basis for diagnosing physical inconsistencies and tracking progress toward physically consistent video world models.

cs.CV↗
arXiv2026

PPTBench: Can Coding Agents Reconstruct the Visual World through Structured, Editable Slides

Coding agents are increasingly moving beyond text-based software tasks to reconstruct visual targets through code. This capability, visual coding, requires agents to translate their understanding of visual targets into executable code. Slides provide a natural testbed for this capability, combining rich visual structure with objects that can be programmatically created and edited. To measure this capability, we introduce PPTBench, a benchmark for reconstructing scientific flow diagrams as editable PowerPoint slides. PPTBench covers 500 scientific flow-diagram tasks across 10 presentation domains, drawn from real research papers, and is evaluated with a four-stage agentic judge covering artifact validity, process and connector fidelity, rendering quality, and visual fidelity. Evaluation of ten models across 36 model--harness--effort configurations shows a substantial gap in reliable visual coding: the best configuration achieves 77.34, while the median across configurations is 24.38. Fine-grained analysis shows that agents can generally produce valid slide files, but struggle to produce high-quality reconstructions that faithfully recover the semantics and visual structure of the target. Further analysis shows that increasing reasoning effort primarily improves hard-gate passage rather than mean detail quality on each configuration's passing tasks, while configurations with more inspection tend to achieve higher overall scores. PPTBench establishes a measurable testbed for studying visual coding and advancing agents toward more reliable visual creation.

cs.CL↗
arXiv2026

Unlearning Is Not Just Erasing: Temporal Decoupling via Generation Inequality

Large language models (LLMs) require effective unlearning to address privacy regulations and safety concerns. However, achieving precise forgetting without compromising general utility remains challenging. Existing sequence- and token-level methods penalize target outputs without modeling their context-dependent retrieval paths, which can disrupt linguistic structure or suppress benign knowledge. We present ADU, a fine-grained, training-based framework that shifts unlearning from token erasure to contextual attention-pathway decoupling. Exploiting the functional distinction between local and global attention heads, ADU identifies preplan positions that retrieve persistent sensitive anchors and fixes their candidate paths under the original model. It then trains attention-projection adapters to suppress attention mass along these paths while preserving local-attention structure and retain-set language modeling. Post-training activation exchange tests whether the modified attention-output module transmits the learned forgetting effect. ADU achieves the strongest aggregate performance among evaluated baselines on the TOFU and WMDP benchmarks, including a Forget Quality of (0.93) on TOFU. It preserves 87--98% of model utility (92.9% on average versus 81.9% for baselines) while reducing side effects in benign contexts.

cs.CL↗
arXiv2026

SWE Refactor Bench: Can Coding Agents Complete a Long-Horizon, Whole-Repository Stack Migration?

Modern software systems accumulate technical debt over decades of development, which makes migration expensive and largely manual. As coding agents become increasingly capable at bug fixing, can they autonomously perform such migrations? Existing benchmarks cannot answer this question because they evaluate only behavioural correctness, not whether the migration actually occurred. This leads an easy hack: agents copy the original implementation to make tests pass. We call this Blindness. To address this problem, we introduce SWE Refactor Bench, a benchmark comprising 20 whole-repository migrations, covering 4 kinds of technical debt. A three-stage evaluation protocol measures both migration completeness and behavioural correctness. (1) Migration Audit verifies that the migration occurred. (2) Behavioural Tests measure correctness with a fixed test suite. (3) Agentic Verification uses 6 independent coding agents to generate targeted tests for hidden behavioural differences. Across 520 runs from 8 frontier models and 26 model-effort configurations, only 28 of 520 runs ($5.4\%$) pass all three stages, 13 of the 20 tasks receive no accepted solution, and the best model (claude-opus-5) scores $47.0/100$. Migration completeness and behavioural correctness are distinct abilities: a few runs preserve behaviour by skipping the migration and are stopped at Migration Audit; most attempt it and break behaviour, and are stopped at Behavioural Tests. Agents cannot deliver a perfect migration: among the 340 runs that pass Migration Audit, $58\%$ reach $99\%$ of the fixed checks, yet only $26\%$ reach $100\%$. Agent capability differs across migration categories: agents score $31.4$ on build toolchain rewrites but only $5.6$ on language rewrites. Together, these findings position SWE Refactor Bench as a rigorous testbed for developing coding agents for reliable whole-repository migrations.

cs.CL↗
arXiv2026

AI4AI-Bench: Benchmarking LLM Agents in Algorithmic Design for Recursive Self-Improvement

Recursive self-improvement (RSI) asks whether an AI system can improve the process that produces AI systems, so that the next system inherits the improvement. That process is the training algorithm: a better objective or update rule improves the compute\mbox{-}capability exchange rate for every subsequent run, including the one that produces the next agent. Whether RSI is feasible therefore turns on whether an agent can design training algorithms. No benchmark isolates that ability: existing suites are won by collecting data or by tuning hyperparameters, and none tells a change to how a run is executed apart from a change to how the model learns. We present AI4AI\mbox{-}Bench, 10 frozen research repositories spanning 10 training algorithm families. In each task, an agent has 4 hours on one B300 to rewrite the training algorithm; its code is then rerun from scratch for up to 12 hours and scored by a fixed evaluator hidden from the agent, against the repository's original algorithm under the same procedure. Because the 10 metrics are incommensurable, every task is mapped onto one scale on which $0$ is an uninformative model, $0.1$ is the algorithm the repository ships, and $1.0$ is the task optimum. Across 29 configurations of 6 systems on all 10 tasks the mean score is $0.166$, and the best system reaches $0.250$: even the strongest closes under a fifth of the distance between the algorithm that was already there and the optimum. The submissions show where that distance went: most never change how the model learns at all, and the minority that do average $0.226$ against $0.126$ for the rest. More reasoning effort mostly buys the willingness to go there, taking that minority from $8\%$ of submissions to $64\%$ and the mean score from $0.094$ to $0.196$. We release the task suite, the evaluators and every scored submission, so that the measurement can be repeated as these systems change.

cs.AI↗
arXiv2026

Agogic: Performance-Timed Music Tokens for LLM-Native Text-to-Symbolic-Music Generation

Text-to-music language models begin with a choice usually made by default: how to tokenize music. Normally entangled with backbone, data, and recipe, its effect has never been measured in isolation. We fix pretrained Qwen3.5 (0.8B-27B), data, budget, and decoding, and swap only the representation across seven tokenizations, anchoring texture metrics to each representation's model-free ceiling. The ordering is clean and surprising: representation, not model size, is the binding variable for distributional fidelity. Scaling the backbone 34x barely moves Frechet Music Distance (FMD), whereas switching representation halves it. PMT, a performance-resolution stream we release (10 ms timing, per-note velocity, multi-track texture; 609 symbols), reaches FMD 159 at 0.8B against 272-286 for beat grids (1.7-1.8x lower, up to 2.8x elsewhere; non-overlapping bootstrap CIs), so a 0.8B performance-resolution model beats a 27B beat grid. It reappears on a 26M from-scratch backbone and a second performance-resolution tokenizer: a property of the class, not one lucky vocabulary. Nor is it a finer-lattice artifact: snapping PMT's onsets to the beat grids' resolution still leaves it 67-129 FMD ahead of both (n=500). The effect is distributional; whether it is audible is a separate question, left open by our probe, with a human study pre-registered. Native caption adherence is weak but separable: a lightweight decode-time constraint doubles instrument-F1 (.28 to .60) and Correct-Key (.16 to .35) at no distributional cost. We release the harness, 25+ checkpoints, two corpora (86.6k aligned across caption/MIDI/ABC/audio; 6.25M captioned, the largest for music), and an imprinting diagnostic: published text-to-MIDI systems reproduce their training distribution near-invariant to the caption (72% vs. 71% chord-time on disjoint domains). The field's next representation claim can now be measured, not asserted.

cs.SD↗
arXiv2026

AdapShot: Adaptive Many-Shot In-Context Learning with Semantic-Aware KV Cache Reuse

Many-Shot In-Context Learning (ICL) has emerged as a promising paradigm, leveraging extensive examples to unlock the reasoning potential of Large Language Models (LLMs). However, existing methods typically rely on a predetermined, fixed number of shots. This static approach often fails to adapt to the varying difficulty of different queries, leading to either insufficient context or interference from noise. Furthermore, the prohibitive computational and memory costs of long contexts severely limit Many-Shot's feasibility. To address the above limitations, we propose AdapShot, which dynamically optimizes shot counts and leverages KV cache reuse for efficient inference. Specifically, we design a probe-based evaluation mechanism that utilizes output entropy to determine the optimal number of shots. To bypass the redundant prefilling computation during both the probing and inference phases, we incorporate a semantics-aware KV cache reuse strategy. Within this reuse strategy, to address positional encoding incompatibilities, we introduce a decoupling and re-encoding method that enables the flexible reordering of cached key-value pairs. Extensive experiments demonstrate that AdapShot achieves an average performance gain of around 10% and a 4.64x speedup compared to state-of-the-art DBSA.

cs.AI↗
arXiv2026

CAP: Controllable Alignment Prompting for Unlearning in LLMs

Large language models (LLMs) trained on unfiltered corpora inherently risk retaining sensitive information, necessitating selective knowledge unlearning for regulatory compliance and ethical safety. However, existing parameter-modifying methods face fundamental limitations: high computational costs, uncontrollable forgetting boundaries, and strict dependency on model weight access. These constraints render them impractical for closed-source models, yet current non-invasive alternatives remain unsystematic and reliant on empirical experience. To address these challenges, we propose the Controllable Alignment Prompting for Unlearning (CAP) framework, an end-to-end prompt-driven unlearning paradigm. CAP decouples unlearning into a learnable prompt optimization process via reinforcement learning, where a prompt generator collaborates with the LLM to suppress target knowledge while preserving general capabilities selectively. This approach enables reversible knowledge restoration through prompt revocation. Extensive experiments demonstrate that CAP achieves precise, controllable unlearning without updating model parameters, establishing a dynamic alignment mechanism that overcomes the transferability limitations of prior methods.

cs.LG↗
arXiv2026

Relit-LiVE: Relight Video by Jointly Learning Environment Video

Recent advances have shown that large-scale video diffusion models can be repurposed as neural renderers by first decomposing videos into intrinsic scene representations and then performing forward rendering under novel illumination. While promising, this paradigm fundamentally relies on accurate intrinsic decomposition, which remains highly unreliable for real-world videos and often leads to distorted appearances, broken materials, and accumulated temporal artifacts during relighting. In this work, we present Relit-LiVE, a novel video relighting framework that produces physically consistent, temporally stable results without requiring prior knowledge of camera pose. Our key insight is to explicitly introduce raw reference images into the rendering process, enabling the model to recover critical scene cues that are inevitably lost or corrupted in intrinsic representations. Furthermore, we propose a novel environment video prediction formulation that simultaneously generates relit videos and per-frame environment maps aligned with each camera viewpoint in a single diffusion process. This joint prediction enforces strong geometric-illumination alignment and naturally supports dynamic lighting and camera motion, significantly improving physical consistency in video relighting while easing the requirement of known per-frame camera pose. Extensive experiments demonstrate that Relit-LiVE consistently outperforms state-of-the-art video relighting and neural rendering methods across synthetic and real-world benchmarks. Beyond relighting, our framework naturally supports a wide range of downstream applications, including scene-level rendering, material editing, object insertion, and streaming video relighting. The Project is available at https://github.com/zhuxing0/Relit-LiVE.

cs.CV↗
arXiv2026

PAM: A Pose-Appearance-Motion Engine for Sim-to-Real HOI Video Generation

Hand-object interaction (HOI) reconstruction and synthesis are becoming central to embodied AI and AR/VR. Yet, despite rapid progress, existing HOI generation research remains fragmented across three disjoint tracks: (1) pose-only synthesis that predicts MANO trajectories without producing pixels; (2) single-image HOI generation that hallucinates appearance from masks or 2D cues but lacks dynamics; and (3) video generation methods that require both the entire pose sequence and the ground-truth first frame as inputs, preventing true sim-to-real deployment. Inspired by the philosophy of Joo et al. (2018), we think that HOI generation requires a unified engine that brings together pose, appearance, and motion within one coherent framework. Thus we introduce PAM: a Pose-Appearance-Motion Engine for controllable HOI video generation. The performance of our engine is validated by: (1) On DexYCB, we obtain an FVD of 29.13 (vs. 38.83 for InterDyn), and MPJPE of 19.37 mm (vs. 30.05 mm for CosHand), while generating higher-resolution 480x720 videos compared to 256x256 and 256x384 baselines. (2) On OAKINK2, our full multi-condition model improves FVD from 68.76 to 46.31. (3) An ablation over input conditions on DexYCB shows that combining depth, segmentation, and keypoints consistently yields the best results. (4) For a downstream hand pose estimation task using SimpleHand, augmenting training with 3,400 synthetic videos (207k frames) allows a model trained on only 50% of the real data plus our synthetic data to match the 100% real baseline.

cs.CV↗
arXiv2026

Benchmarking PhD-Level Coding in 3D Geometric Computer Vision

AI-assisted coding has rapidly reshaped software practice and research workflows, yet today's models still struggle to produce correct code for complex 3D geometric vision. If models could reliably write such code, the research of our community would change substantially. To measure progress toward that goal, we introduce GeoCodeBench, a PhD-level benchmark that evaluates coding for 3D vision. Each problem is a fill-in-the-function implementation task curated from representative papers at recent venues: we first let a tool propose candidate functions from official repositories, then perform careful human screening to select core 3D geometric components. For every target, we generate diverse, edge-case unit tests, enabling fully automatic, reproducible scoring. We evaluate eight representative open- and closed-source models to reflect the current ecosystem. The best model, GPT-5, attains only 36.6% pass rate, revealing a large gap between current capabilities and dependable 3D scientific coding. GeoCodeBench organizes tasks into a two-level hierarchy: General 3D capability (geometric transformations and mechanics/optics formulation) and Research capability (novel algorithm implementation and geometric logic routing). Scores are positively correlated across these axes, but research-oriented tasks are markedly harder. Context ablations further show that "more paper text" is not always better: cutting off at the Method section statistically outperforms full-paper inputs, highlighting unresolved challenges in long-context scientific comprehension. Together, these findings position GeoCodeBench as a rigorous testbed for advancing from generic coding to trustworthy 3D geometric vision coding.

cs.CV↗
arXiv2025

Zoom in, Click out: Unlocking and Evaluating the Potential of Zooming for GUI Grounding

Grounding is a fundamental capability for building graphical user interface (GUI) agents. Although existing approaches rely on large-scale bounding box supervision, they still face various challenges, such as cross-platform generalization, complex layout analysis, and fine-grained element localization. In this paper, we investigate zoom as a strong yet underexplored prior for GUI grounding, and propose a training-free method, ZoomClick. By characterizing four key properties of zoom (i.e., pre-zoom, depth, shrink size, minimal crop size), we unlock its full capabilities for dynamic spatial focusing and adaptive context switching. Experiments demonstrate that our method significantly boosts the performance of both general vision-language and specialized GUI grounding models, achieving state-of-the-art results on several mainstream benchmarks; for example, UI-Venus-72B attains a 73.1% success rate on ScreenSpot-Pro. Furthermore, we present GUIZoom-Bench, a benchmark for evaluating model adaptability to zoom, aiming to inspire future research on improving zoom for further training and test-time scaling in GUI grounding tasks.

cs.CV↗
arXiv2025

SCP-Diff: Spatial-Categorical Joint Prior for Diffusion Based Semantic Image Synthesis

Semantic image synthesis (SIS) shows good promises for sensor simulation. However, current best practices in this field, based on GANs, have not yet reached the desired level of quality. As latent diffusion models make significant strides in image generation, we are prompted to evaluate ControlNet, a notable method for its dense control capabilities. Our investigation uncovered two primary issues with its results: the presence of weird sub-structures within large semantic areas and the misalignment of content with the semantic mask. Through empirical study, we pinpointed the cause of these problems as a mismatch between the noised training data distribution and the standard normal prior applied at the inference stage. To address this challenge, we developed specific noise priors for SIS, encompassing spatial, categorical, and a novel spatial-categorical joint prior for inference. This approach, which we have named SCP-Diff, has set new state-of-the-art results in SIS on Cityscapes, ADE20K and COCO-Stuff, yielding a FID as low as 10.53 on Cityscapes. The code and models can be accessed via the project page.

cs.CV↗
arXiv2025

Impromptu VLA: Open Weights and Open Data for Driving Vision-Language-Action Models

Vision-Language-Action (VLA) models for autonomous driving show promise but falter in unstructured corner case scenarios, largely due to a scarcity of targeted benchmarks. To address this, we introduce Impromptu VLA. Our core contribution is the Impromptu VLA Dataset: over 80,000 meticulously curated video clips, distilled from over 2M source clips sourced from 8 open-source large-scale datasets. This dataset is built upon our novel taxonomy of four challenging unstructured categories and features rich, planning-oriented question-answering annotations and action trajectories. Crucially, experiments demonstrate that VLAs trained with our dataset achieve substantial performance gains on established benchmarks--improving closed-loop NeuroNCAP scores and collision rates, and reaching near state-of-the-art L2 accuracy in open-loop nuScenes trajectory prediction. Furthermore, our Q&A suite serves as an effective diagnostic, revealing clear VLM improvements in perception, prediction, and planning. Our code, data and models are available at https://github.com/ahydchh/Impromptu-VLA.

cs.CV↗
arXiv2025

FB-4D: Spatial-Temporal Coherent Dynamic 3D Content Generation with Feature Banks

With the rapid advancements in diffusion models and 3D generation techniques, dynamic 3D content generation has become a crucial research area. However, achieving high-fidelity 4D (dynamic 3D) generation with strong spatial-temporal consistency remains a challenging task. Inspired by recent findings that pretrained diffusion features capture rich correspondences, we propose FB-4D, a novel 4D generation framework that integrates a Feature Bank mechanism to enhance both spatial and temporal consistency in generated frames. In FB-4D, we store features extracted from previous frames and fuse them into the process of generating subsequent frames, ensuring consistent characteristics across both time and multiple views. To ensure a compact representation, the Feature Bank is updated by a proposed dynamic merging mechanism. Leveraging this Feature Bank, we demonstrate for the first time that generating additional reference sequences through multiple autoregressive iterations can continuously improve generation performance. Experimental results show that FB-4D significantly outperforms existing methods in terms of rendering quality, spatial-temporal consistency, and robustness. It surpasses all multi-view generation tuning-free approaches by a large margin and achieves performance on par with training-based methods.

cs.CV↗