Search arXiv⌕ Search

arXiv subjects

Zeyu Wang

Publications and source records attributed to Zeyu Wang.

At least 37 records · Page 2Linked to original sources
arXiv2026

CARE: Context-Aware Ranking Evolution with Executable Scoring Programs for Budgeted Reaction Optimization

High-throughput experimentation can evaluate many reaction conditions, yet combinatorial condition spaces still exceed the available experiment budget. This makes experiment selection a sequential decision problem: each new condition must be chosen from limited observations before its outcome is known. LLMs can express task-specific selection logic. A direct recommendation, however, is neither a persistent executable object that can be validated and revised nor an independently auditable decision rule. We introduce CARE, a reference-conditioned controller that separates program synthesis from experiment selection. An LLM writes an executable scoring program that ranks the remaining conditions, while a non-LLM reference policy supplies a numerical candidate and support summary. CARE forms an optional alternative from the program, applies a reference-conditioned intervention gate to compare it with the reference, and records the decision before the selected outcome is revealed. Each new outcome updates controller state and can trigger retention, revision, or regeneration of the active program. This outcome-guided program evolution changes the scoring logic without updating the LLM parameters. In matched offline replay with 30 seeds on eight reaction-optimization tasks, CARE attains the lowest normalized regret, the highest normalized best-so-far AUC, and the highest Top-1% Success@15 among the evaluated methods. These results support using a scoring program written by an LLM as one component of a reference-conditioned optimizer rather than as a standalone experiment selector.

cs.LG↗
arXiv2026

MVID: Feed-Forward Multi-View Intrinsic Image Decomposition

Intrinsic image decomposition aims to recover material and illumination factors from RGB observations, but real-world images entangle reflectance with illumination, visibility, shadows, and non-diffuse appearance. Recent single-image methods address this entanglement with a residual image formation model, decomposing RGB into albedo, diffuse shading, and a non-diffuse residual. However, applying such decomposition independently to consecutive frames lacks scene-level context, leading to inconsistent factor assignments, albedo drift, and leakage across views. Meanwhile, multi-view inverse-rendering methods recover properties, such as material, lighting, and geometry, but they rely on synthetic data due to the highly uncertain estimation, and do not generalize to supervision by real images. We present MVID, Multi-View Intrinsic image Decomposition, a feed-forward framework built on the residual image formation model. MVID builds a scene-level multi-view representation and decodes a view-consistent albedo together with coherent per-view shading and residual factors through a factor query adapter, while using the same image formation model for self-supervised RGB reconstruction on unlabeled real-world sequences. Experiments on indoor, real-world, and outdoor benchmarks show that MVID improves both per-frame decomposition quality and cross-view consistency over single-view intrinsic, generative intrinsic, and multi-view inverse-rendering baselines. The resulting view-stable factors support practical image-space applications, including multi-view consistent illumination editing and specularity removal.

cs.CV↗
arXiv2026

Thinking Beyond Videos: Unifying Video Reasoning and Deep Research for Open-World Video Agents

Open-world video understanding often requires a model to locate sparse visual evidence and acquire external knowledge that is absent from the video and its parametric memory. While Thinking-with-Videos enables active temporal perception and Deep Research supports multi-step information seeking, the two capabilities are typically developed in isolation. We introduce VideoRover, a unified Video Deep Research framework that iteratively coordinates video cropping, multimodal search, and webpage browsing. Given a video-question pair, VideoRover uses each tool result to select the next action, so localized video clips guide external retrieval and retrieved evidence triggers further video inspection and verification. To develop this capability, we construct an automated data curation pipeline, producing 26K verified SFT trajectories and 3K challenging RL instances. We also introduce VideoRover-Bench, a benchmark stratified by video duration and research difficulty. Experiments on VideoDR and VideoRover-Bench show that our VideoRover-8B-RL achieves performance comparable to proprietary models in the direct-answer setting without tool use while outperforming larger open-source models equipped with the same tool suite. Ablation studies and training dynamics further validate the complementary roles of active video grounding, external retrieval, and long-horizon reinforcement learning.

cs.CV↗
arXiv2026

Knowing Isn't Always Saying: When Do Spatial Encodings Reach Answers in Vision-Language Models?

Vision-language models are known to encode spatial information in their hidden states, yet often fail to use it when answering. However, it remains unclear when and where this encoded information reaches the answer. We address this with direction patching, a class-conditioned causal intervention applied across layers, token positions, and prompt formats. Using spatial-ID directions constructed following prior encoding evidence, we find that causal influence on answer logits emerges only at mid-to-deep depths. Text chain-of-thought suppresses immediate object-word argmax-level transport in most models, while visually grounded prompts keep it open. Positive target-logit gain can remain below the argmax threshold, and transport can re-emerge at the final prefix token or at the answer step in deeper layers. Across the ten VLMs we study, these local effects form descriptive transport patterns. Complementary experiments characterize how these patterns shift across datasets, attributes, and encoding amplitudes. Together, these results reframe the encoding-grounding gap as a problem of conditional transport in VLMs.

cs.CL↗
arXiv2026

TopoGS: Topology-Aware Anchor Feature Aggregation for Large-Scale 3D Gaussian Splatting

Octree-based 3D Gaussian Splatting organizes anchors into multi-level hierarchies for level-of-detail rendering, but features at different levels are typically optimized independently, leaving the octree topology underused during feature learning. We observe that uniform cross-level aggregation produces asymmetric effects: fine-level anchors benefit from coarse context, whereas coarse-level anchors require selective information from their descendants. We therefore propose TopoGS, a topology-aware anchor feature aggregation framework with two lightweight components. Hierarchical Anchor Coupling establishes bidirectional cross-level gradient pathways by fusing per-level context triplets with a residual MLP. Structure-Aware Containment Aggregation uses octree containment and hash-based matching to distinguish anchors with valid parent-child relations from isolated anchors, then applies soft weighting to accommodate varying topological sparsity. Experiments on ten scenes from Mill19, UrbanScene3D, Tanks & Temples, MatrixCity, and WHU show consistent improvements over state-of-the-art methods. TopoGS achieves average PSNR gains of 2.13, 1.78, and 0.29 dB over the strongest reported baseline on aerial, ground-level, and synthetic-cartographic scenes, respectively, while rendering faster and using less memory. Code is available at https://github.com/WZ-CS/TopoGS.

cs.CV↗
arXiv2026

MSEditor: Toward Consistent Multi-Shot Video Editing

In this paper, we tackle the problem of performing consistent, unified modifications to a multi-shot video sequence. This task is particularly challenging because multi-shot videos consist of discontinuous temporal segments that vary significantly in viewpoint, camera scale, and subject pose, leading to severe identity drift and cumulative error propagation. Achieving coherent edits requires establishing reliable cross-shot semantic awareness to maintain stable subject appearance and visual continuity across these disjointed boundaries. To address this, we propose MSEditor, the first framework designed specifically for consistent multi-shot video editing. To overcome the scarcity of high-quality multi-shot training data, we repurpose existing multi-view video datasets to provide robust cross-shot supervision. Architecturally, we introduce a Supervisory Adapter that injects this cross-shot information into the diffusion backbone, enabling the model to learn identity-consistent representations. Furthermore, to effectively mitigate cumulative errors and ensure long-range temporal coherence, we design a Cross-Shot Packing strategy that dynamically aggregates information from semantically related shots within the self-attention window. Extensive experiments demonstrate that MSEditor significantly outperforms existing methods on our curated multi-shot video editing benchmark in terms of identity preservation, temporal stability, and overall visual quality.

cs.CV↗
arXiv2026

HLE-Verified: A Systematic Verification and Structured Revision of Humanity's Last Exam

Humanity's Last Exam (HLE) has become a widely used benchmark for evaluating frontier large language models on challenging, multi-domain questions. However, community-led analyses have raised concerns that HLE contains a non-trivial number of noisy items, which can bias evaluation results and distort cross-model comparisons. To address this challenge, we introduce HLE-Verified, a verified and revised version of HLE with a transparent verification protocol and fine-grained error taxonomy. Our construction follows a two-stage validation-and-repair workflow resulting in a certified benchmark. In Stage I, each item undergoes binary validation of the problem and final answer through domain-expert review and model-based cross-checks, yielding 668 verified items. In Stage II, flawed but fixable items are revised under strict constraints preserving the original evaluation intent, through dual independent expert repairs, model-assisted auditing, and final adjudication, resulting in 1,143 revised-and-certified items. The remaining 689 items are released as a documented uncertain set with explicit uncertainty sources and expertise tags for future refinement. We evaluate eight state-of-the-art language models on HLE and HLE-Verified, observing an average absolute accuracy gain of 7--10 percentage points on HLE-Verified. The improvement is particularly pronounced on items where the original problem statement and/or reference answer is erroneous, with gains of 30--40 percentage points. Our analyses further reveal a strong association between model confidence and the presence of errors in the problem statement or reference answer, supporting the effectiveness of our revisions. Overall, HLE-Verified improves HLE-style evaluations by reducing annotation noise and enabling more faithful measurement of model capabilities. Data is available at: https://huggingface.co/datasets/skylenage/HLE-Verified

cs.CL↗
arXiv2026

An $h$-adaptive Tetrahedral Spectral Element Method with Applications to Kohn-Sham Density Functional Theory

High-order $h$-adaptive spectral element methods on tetrahedral meshes provide an effective framework for resolving localized singularities and multiscale structures in complex three-dimensional geometries. However, their development is often hindered by difficulties in maintaining $C^0$ continuity across refinement interfaces and efficiently transferring solutions between adaptive meshes. Such limitations are particularly relevant in demanding applications such as all-electron Kohn-Sham density functional theory, which place stringent requirements on the accurate resolution of both nuclear singularities and multiple physical scales. In this paper, we present an efficient $h$-adaptive tetrahedral spectral element framework. To address the continuity challenge, we develop an adaptive strategy that combines element orientation alignment with geometric red-green refinement, thereby eliminating the need for algebraic hanging-node constraints while preserving inter-element continuity. Furthermore, an efficient topology-based point-location algorithm is introduced to accelerate interpolation between adaptive meshes. Numerical experiments on Poisson and Laplacian eigenvalue problems confirm the spectral convergence of the proposed method. Applications to all-electron Kohn-Sham equations further demonstrate its capability to accurately resolve nuclear singularities. Moreover, parallel performance studies exhibit excellent scalability, with matrix assembly and adaptivity modules generally achieving speedups above 15 times and the proposed interpolation algorithm attaining speedups ranging from 25 to 35 on 64-core configurations compared to the single-core performance. These results indicate that the proposed framework provides an accurate, robust, and efficient solution for large-scale, high-resolution simulations.

math.NA↗
arXiv2026

MERA: Model Evolution and Routing with Skill Adaptation for Agentic Systems at Scale

LLM agents execute heterogeneous sequences of model calls within a single task: some invocations require careful reasoning, while others are structured steps such as formatting or tool-argument construction. Prior routing methods exploit this asymmetry by assigning easy invocations to a cheaper small model and difficult ones to a large model. Such policies reduce inference cost, but they leave the small model's capability unchanged, so attainable savings remain bounded by the work the student can already solve. MERA instead improves the small model itself, using a single model invocation as the unit of adaptation. In each cycle, MERA replays failed student invocations to obtain execution-verified teacher demonstrations, distills recurring procedures into an iteratively updated SkillBook, and fine-tunes a student LoRA adapter via supervised learning and optional GRPO. Routing serves as supporting machinery for deployment: the improved student is served behind a cost-calibrated router with verifier-backed fallback, and a candidate SkillBook, adapter, or router is admitted only when joint replay preserves task quality. Empirically, four-cycle adaptation raises Qwen2.5-Coder-1.5B from 28.7% to 49.7% pass on held-out HumanEval+MBPP. Under verifier-backed fallback, the deployed policy retains 88.3% pass at 60.8% of always-Luna cost. On TAU-2, a fine-tuned Qwen3.5-2B improves from 14/35 to 18/35 and matches an unadapted 4B model. These results indicate that verifier-backed multi-cycle adaptation can increase small-model capability, rather than only routing around a fixed student.

cs.LG↗
arXiv2026

EmoWorld: A Decoupled Affective Field for Controllable Emotional Video Generation

Emotion shapes how viewers interpret a scene, yet existing video generators entangle global atmosphere, affect-bearing semantic cues, and temporal progression within a single text condition. We present EmoWorld, a framework that decouples these factors within a frozen flow-matching video diffusion transformer (Video DiT). A one-time preparation stage extracts layer-specific affect directions and a reusable cue library from geometry-preserving neutral and emotion-edited panoramas. At inference, Visual Atmosphere Steering (VAS) injects atmosphere directions into hidden states, Semantic Affective Steering (SAS) isolates a separately scalable prompt residual for semantic cues, and Temporal Affective Steering (TAS) interpolates endpoint residual fields across denoising and video time. On Wan2.2, VAS improves target-emotion alignment by 19% while reducing a temporal-fluctuation proxy by 48%; SAS improves target-emotion alignment by 37% and increases detected affect-bearing cues by 36%; and TAS improves transition monotonicity by 15% over the strongest baseline. EmoWorld is evaluated across 27 emotion categories in text-to-video and image-to-video settings, demonstrates portability across multiple Video-DiT backbones, and supports camera-conditioned composition without updating generator parameters.

cs.CV↗
arXiv2026

Decoupling Generation and Selection for Budget-Constrained Faithful Summarization

Abstractive summarization models remain vulnerable to factual inconsistency, redundancy, and weak length control. We propose a modular generation-and-selection framework for sentence-budget-constrained summarization. A pretrained generator produces multiple candidate summaries, which are decomposed into sentence-level candidates. A combinatorial selector then constructs the final summary by balancing relevance, factuality, and redundancy under an explicit budget. The framework supports MMR, ILP, and a DPP-inspired log-determinant objective without retraining the generator. Experiments on CNN/DailyMail, Multi-News, FaithBench, and TofuEval show consistent improvements in factuality and source-grounding metrics, especially for multi-document summarization, at the cost of lower reference-overlap scores. Human evaluation further indicates higher perceived consistency, relevance, clarity, and conciseness, with a small reduction in coherence. These results show that decoupling generation from selection provides a model-agnostic mechanism for improving factual grounding. Code is available at https://anonymous.4open.science/r/bcfs-D05E/.

cs.CL↗
arXiv2026

Towards Wearable Opportunistic Crowdsensing for Open-Vocabulary Activity Data Collection Through User-Scheduled Trigger-Action Routines

Collecting richly labeled wearable activity data in everyday settings remains difficult because retrospective annotation is costly and often imprecise. Prior data collection apps rely on a labor-intensive self-reporting strategy and primarily treat participants as crowd labelers. We present Pebbl, a feasibility-stage system that incentivizes in-situ labeling through opportunistic crowdsensing. Pebbl lets users author trigger-action recipes on a smartphone and receive just-in-time reminders for beneficial actions when a trigger is detected. In the prototype, triggers are a limited set with four common audio cues, while actions are described in open-vocabulary natural language. Each confirmed execution yields a short sensor window with explicit start/end boundaries and a user-authored action label. We evaluate Pebbl through an expert workshop with wearable Human Activity Recognition (HAR) researchers (N = 6), a within-subject in-lab study (N = 21), and a pilot deployment (N = 8). Experts viewed the approach as lower burden and more ecologically valid than common labeling workflows. In the lab, Pebbl produced reliable execution logs under controlled conditions (recall = 97.30%, precision = 97.15%) and was preferred over comparison workflows on perceived burden and confidence. The pilot deployment shows that the interaction and sensing pipeline can function in free-living use, while surfacing practical constraints such as false triggers and context dependence. Overall, Pebbl represents a step toward a low-burden, distributable collection approach of user-contributed wearable activity data.

cs.HC↗
arXiv2026

Collascope: Supporting Serendipitous Asset Exploration for Collage-Based Storytelling

Collage-based storytelling requires visual elements that support emerging narratives and inspire creative reinterpretation. Existing tools, however, rely largely on keyword- and image-based retrieval, offering limited support for serendipitous exploration beyond existing assets. We introduce Collascope, an interactive system that helps creators (1) concretize story intent with interactive element groups, (2) expand the exploration space based on concepts or cutouts towards conceptual and visual dimensions, and (3) develop grounded, traceable ideas in parallel with collage composition. Collascope's attribute-aware visual retrieval method, instantiated with collage-relevant visual dimensions, enables creators to retrieve cutouts through dimension-specific visual projections rather than holistic similarity. In a within-subject study (N=12) against a conventional search baseline, our participants used unexpected results and even gaps in the asset collection to redirect narratives, shift tone, and enrich compositions. Scene Parts helped organize exploration into manageable subtasks, while participants used association in distinct ways depending on whether exploration was guided by a clear goal, an evolving story, or visual intuition.

cs.HC↗
arXiv2026

The Sparsity Ceiling: Where Spiking Networks Can and Cannot Trade Activity for Energy

Spiking neural networks (SNNs) are promoted as an energy-efficient substrate because sparse, event-driven activity replaces dense multiply-accumulates with cheap accumulates. We argue the energy dividend of sparsity is not a property of SNNs but of the task. Holding architecture fixed and swapping only the hidden unit (continuous vs. leaky-integrate-and-fire), plus a two-sided target-firing-rate probe, we measure how far activity can be pushed down before quality breaks. Low-load feed-forward perception sparsifies to 5% firing at no accuracy cost; a recurrent language model cannot go below ~50% -- the recurrent state must stay active to carry information. A spiking Transformer, by contrast, sparsifies freely to 2% (3 seeds) -- so the ceiling is a property of recurrent compression, not sequence modeling. Attention escapes the floor only by storing the full key-value cache, trading a firing floor for a memory wall: on neuromorphic hardware, recurrence and attention pay on different axes, neither escapes. We formalize the ceiling with an information-theoretic bound rho >= H_b^{-1}(log2 M / H) and confirm its predictions: the floor rises with memory load, falls with state width, and (refuting a naive memory-only reading) rises with task difficulty. A layer-wise input floor further caps op reduction under dense input, isolating event-driven perception as where neuromorphic hardware wins.

cs.NE↗
arXiv2026

Spatula: Exploring On-Demand In-Situ Interfaces and Interaction for Attribute Control

Controlling attributes is a critical step toward achieving the final creative outcome, yet current approaches fall short in supporting users in the iterative refinement of generative content. We propose Spatula, a proof-of-concept system that generates on-demand, in-situ attribute control interfaces and interactions for creating motion graphics. Building on a technical probe that automatically analyzes animation context and generates corresponding attributes and UI, we frame attribute control as an explorable landscape and explore the attribute control space along four key dimensions: Discoverability, Resolution, Scope, and Expandability. Findings from a user study (N=12) show that our system provides intuitive and convenient interactions while supporting diverse needs for fine-grained parameter control. Furthermore, our applications demonstrate that the plug-and-play design generalizes to other domains, such as web design and 3D modeling.

cs.HC↗
arXiv2026

Glass Surface Detection: Leveraging Reflection Dynamics in Flash/No-flash Imagery

Glass surfaces are ubiquitous in daily life, typically appearing colorless, transparent, and lacking distinctive features. These characteristics make glass surface detection a challenging computer vision task. Existing glass surface detection methods always rely on boundary cues (\textit{e.g.}, window and door frames) or reflection cues to locate glass surfaces, but they fail to fully exploit the intrinsic properties of the glass itself for accurate localization. We observed that in most real-world scenes, the illumination intensity in front of the glass surface differs from that behind it, which results in variations in the reflections visible on the glass surface. Specifically, when standing on the brighter side of the glass and applying a flash towards the darker side, existing reflections on the glass surface tend to disappear. Conversely, while standing on the darker side and applying a flash towards the brighter side, distinct reflections will appear on the glass surface. Based on this phenomenon, we propose \textit{NFGlassNet}, a novel method for glass surface detection that leverages the reflection dynamics present in flash/no-flash imagery. Specifically, we propose a Reflection Contrast Mining Module (RCMM) for extracting reflections, and a Reflection Guided Attention Module (RGAM) for fusing features from reflection and glass surface for accurate glass surface detection. For learning our network, we also construct a dataset consisting of $\sim$3.3$K$ no-flash and flash image pairs captured from various scenes with corresponding ground truth annotations. Extensive experiments demonstrate that our method outperforms the state-of-the-art methods. Our code, model, and dataset will be available upon acceptance of the manuscript.

cs.CV↗
arXiv2026

EmoSpace: Immersive Affective Image Generation Guided by Fine-Grained Emotion Prototypes

Immersive affective content generation aims to create visually compelling VR imagery with controllable emotional nuance, yet existing methods typically rely on coarse labels or prompt-only control. Although modern diffusion transformers (DiTs) such as FLUX improve visual fidelity, they are not designed to incorporate structured affective representations. We present EmoSpace, an immersive affective content generation framework guided by fine-grained emotion prototypes, transforming free-form text and emotion descriptions into fine-grained affective imagery through three coordinated components. First, to represent sub-emotion variation beyond conventional categorical models, EmoSpace learns a hierarchical bank of 256 prototypes with input-conditioned adaptation through vision-language alignment. Second, Prototype-Conditioned Steering converts these prototypes into DiT-compatible generation signals through multi-pathway injection and temporal blending, while Iterative Prompt Refinement enriches prompts with prototype-aligned sub-emotion descriptors. Third, Affect-Grounded Modulation coordinates emotion conditioning with controllable LoRAs for panoramic, stylized, and multi-conditional generation. Through quantitative and qualitative evaluations, EmoSpace improves fine-grained emotional alignment while maintaining high aesthetic quality. Our user study shows that EmoSpace outputs are perceived as more emotionally aligned than baseline results and more suitable for immersive scene design. Additionally, we find that immersive presentation alters emotional perception and increases emotional engagement. Together, these findings inform the design of emotion-aware generative systems for immersive media, with potential applications including education, immersive storytelling, and artistic creation. We will release our code and models to facilitate future research along this line.

cs.CV↗
arXiv2026

ARCANA: A Reflective Multi-Agent Program Synthesis Framework for ARC-AGI-2 Reasoning

We present ARCANA, a collaborative multi agent framework for solving ARC AGI 2 tasks under strict test time and hardware constraints. ARCANA decomposes each task into iterative perception, hypothesis generation, symbolic execution, and reflective refinement. A perceptual grounding agent builds object centric scene graphs from raw grids, a latent program policy proposes diverse DSL programs, a symbolic executor verifies candidates on demonstrations, and a reflective agent synthesizes failure driven feedback for the next turn. These agents communicate through a shared differentiable blackboard and are scheduled by a learned meta controller. The design combines structured program search with adaptive multi turn correction, improving reasoning efficiency and solution quality on challenging abstract transformation tasks.

cs.AI↗