ClawGym: A Scalable Framework for Building Effective Claw Agents

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

Large Language Models Explore by Latent Distilling

GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents

SWE-chat: Coding Agent Interactions From Real Users in the Wild

AdaExplore: Failure-Driven Adaptation and Diversity-Preserving Search for Efficient Kernel Generation

Refinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models

AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery

Meta-CoT: Enhancing Granularity and Generalization in Image Editing

DV-World: Benchmarking Data Visualization Agents in Real-World Scenarios

Programming with Data: Test-Driven Data Engineering for Self-Improving LLMs from Raw Corpora

Recursive Multi-Agent Systems

Skill Retrieval Augmentation for Agentic AI

SketchVLM: Vision language models can annotate images to explain thoughts and guide users

RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-N Ranking

LongSpeech: A Scalable Benchmark for Transcription, Translation and Understanding in Long Speech

ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents

Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation

Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms

ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning

From Skills to Talent: Organising Heterogeneous Agents as a Real-World Company

World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

Video Analysis and Generation via a Semantic Progress Function

SmartPhotoCrafter: Unified Reasoning, Generation and Optimization for Automatic Photographic Image Editing

Contexts are Never Long Enough: Structured Reasoning for Scalable Question Answering over Long Document Sets

AgentSearchBench: A Benchmark for AI Agent Search in the Wild

FlowAnchor: Stabilizing the Editing Signal for Inversion-Free Video Editing

LLM Safety From Within: Detecting Harmful Content with Internal Representations

DiffNR: Diffusion-Enhanced Neural Representation Optimization for Sparse-View 3D Tomographic Reconstruction

Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond

ClawGym: A Scalable Framework for Building Effective Claw Agents

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

Large Language Models Explore by Latent Distilling

GLM-5V-Turbo: Toward a Native Foundation Model for Multimodal Agents

SWE-chat: Coding Agent Interactions From Real Users in the Wild

AdaExplore: Failure-Driven Adaptation and Diversity-Preserving Search for Efficient Kernel Generation

Refinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models

AutoResearchBench: Benchmarking AI Agents on Complex Scientific Literature Discovery

Meta-CoT: Enhancing Granularity and Generalization in Image Editing

DV-World: Benchmarking Data Visualization Agents in Real-World Scenarios

Programming with Data: Test-Driven Data Engineering for Self-Improving LLMs from Raw Corpora

Recursive Multi-Agent Systems

Skill Retrieval Augmentation for Agentic AI

SketchVLM: Vision language models can annotate images to explain thoughts and guide users

RSRCC: A Remote Sensing Regional Change Comprehension Benchmark Constructed via Retrieval-Augmented Best-of-N Ranking

LongSpeech: A Scalable Benchmark for Transcription, Translation and Understanding in Long Speech

ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents

Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation

Vision-Language-Action Safety: Threats, Challenges, Evaluations, and Mechanisms

ReVSI: Rebuilding Visual Spatial Intelligence Evaluation for Accurate Assessment of VLM 3D Reasoning

From Skills to Talent: Organising Heterogeneous Agents as a Real-World Company

World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

Video Analysis and Generation via a Semantic Progress Function

SmartPhotoCrafter: Unified Reasoning, Generation and Optimization for Automatic Photographic Image Editing

Contexts are Never Long Enough: Structured Reasoning for Scalable Question Answering over Long Document Sets

AgentSearchBench: A Benchmark for AI Agent Search in the Wild

FlowAnchor: Stabilizing the Editing Signal for Inversion-Free Video Editing

LLM Safety From Within: Detecting Harmful Content with Internal Representations

DiffNR: Diffusion-Enhanced Neural Representation Optimization for Sparse-View 3D Tomographic Reconstruction

Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond