wovepaper — the arXiv observatory

papers

Publications (67)

cs.CL2026

DeepSearch-World: Self-Distillation for Deep Search Agents in a Verifiable Environment

Xinyu Geng, Xuanhua He, Sixiang Chen +7

The paper introduces DeepSearch-World, a deterministic, verifiable web environment, and DeepSearch-Evolve, a self‑distillation framework that lets web search agents improve from th…

Publications (67)

DeepSearch-World: Self-Distillation for Deep Search Agents in a Verifiable Environment

Code2Math: Can Your Code Agent Effectively Evolve Math Problems Through Exploration?

CultureCLIP: Empowering CLIP with Cultural Awareness through Synthetic Images and Contextualized Captions

The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination

DocCHA: Towards LLM-Augmented Interactive Online diagnosis System

Diversity-Enhanced Reasoning for Subjective Questions

CRAFT: Customizing LLMs by Creating and Retrieving from Specialized Toolsets

EcomBench: Towards Holistic Evaluation of Foundation Agents in E-commerce

Let's Reason Formally: Natural-Formal Hybrid Reasoning Enhances LLM's Math Capability

Enhanced Chart Understanding in Vision and Language Task via Cross-modal Pre-training on Plot Table Pairs

EMCompress: Video-LLMs with Endomorphic Multimodal Compression

MATP-BENCH: Can MLLM Be a Good Automated Theorem Prover for Multimodal Problems?

Alice: Proactive Learning with Teacher's Demonstrations for Weak-to-Strong Generalization

CostBench: Evaluating Multi-Turn Cost-Optimal Planning and Adaptation in Dynamic Environments for LLM Tool-Use Agents

Persona-DB: Efficient Large Language Model Personalization for Response Prediction with Collaborative Data Refinement

UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios

COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation

Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models

Empowering Reliable Visual-Centric Instruction Following in MLLMs

MedEBench: Diagnosing Reliability in Text-Guided Medical Image Editing

CDR-Bench: Evaluating Faithful Execution of Compositional, Order-Sensitive Data Refinement Recipes

MACAROON: Training Vision-Language Models To Be Your Engaged Partners

Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning

LEMMA: Towards LVLM-Enhanced Multimodal Misinformation Detection with External Knowledge Augmentation

VLM2-Bench: A Closer Look at How Well VLMs Implicitly Link Explicit Matching Visual Cues

Video Signature: Implicit Watermarking for Video Diffusion Models

From Pixels to Insights: A Survey on Automatic Chart Understanding in the Era of Large Foundation Models

Supervised Fine-Tuning Needs to Unlock the Potential of Token Priority

Advancing Language Multi-Agent Learning with Credit Re-Assignment for Interactive Environment Generalization

On Stable Long-Form Generation: Benchmarking and Mitigating Length Volatility

CREATOR: Tool Creation for Disentangling Abstract and Concrete Reasoning of Large Language Models

TAMP: Token-Adaptive Layerwise Pruning in Multimodal Large Language Models

NewsClaims: A New Benchmark for Claim Detection from News with Attribute Knowledge

XSkill: Continual Learning from Experience and Skills in Multimodal Agents

Scalable Token-Level Hallucination Detection in Large Language Models

Unveiling the Lack of LVLM Robustness to Fundamental Visual Variations: Why and Path Forward

GeoBrowse: A Geolocation Benchmark for Agentic Tool Use with Expert-Annotated Reasoning Traces

Veri-R1: Toward Precise and Faithful Claim Verification via Online Reinforcement Learning

Defining a New NLP Playground

Mathematical Proof as a Litmus Test: Revealing Failure Modes of Advanced Large Reasoning Models

R-Tuning: Instructing Large Language Models to Say `I Don't Know'

MARS-SQL: A multi-agent reinforcement learning framework for Text-to-SQL

Reinforcement Learning from Denoising Feedback

Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontiers

Decoding the Silent Majority: Inducing Belief Augmented Social Graph with Large Language Model for Response Forecasting

AgentVista: Evaluating Multimodal Agents in Ultra-Challenging Realistic Visual Scenarios

Lean4Physics: Comprehensive Reasoning Framework for College-level Physics in Lean4

RedAct: Redacting Agent Capability Traces for Procedural Skill Protection

Scaling Laws of Synthetic Data for Language Models

CiteGuard: Faithful Citation Attribution for LLMs via Retrieval-Augmented Validation

MimeQA: Towards Socially-Intelligent Nonverbal Foundation Models

NormSAGE: Multi-Lingual Multi-Cultural Norm Discovery from Conversations On-the-Fly

Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning

Towards Self-Evolving Benchmarks: Synthesizing Agent Trajectories via Test-Time Exploration under Validate-by-Reproduce Paradigm

If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents

On the Geometry of On-Policy Distillation

SELF-REDRAFT: Eliciting Intrinsic Exploration-Exploitation Balance in Test-Time Scaling for Code Generation

Are Your Agents Upward Deceivers?

RePPL: Recalibrating Perplexity by Uncertainty in Semantic Propagation and Language Generation for Explainable QA Hallucination Detection

AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting

AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints

A Weibo Dataset for the 2022 Russo-Ukrainian Crisis

SmartBook: AI-Assisted Situation Report Generation for Intelligence Analysts

Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents

MMBoundary: Advancing MLLM Knowledge Boundary Awareness through Reasoning Step Confidence Calibration

Reasoning Path Divergence: A New Metric and Curation Strategy to Unlock LLM Diverse Thinking

Environment Scaling for Interactive Agentic Experience Collection: A Survey