Command Palette
Search for a command to run...
Papers
Daily updated cutting-edge AI research papers to help you keep up with the latest AI trends

LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling

Predicting LLM Safety Before Release by Simulating Deployment































LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling

Predicting LLM Safety Before Release by Simulating Deployment






























FastContext: Training Efficient Repository Explorer for Coding Agents
VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models
DreamX-World 1.0: A General-Purpose Interactive World Model
Geometric Action Model for Robot Policy Learning
Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories
JoyAI-VL-Interaction: Real-Time Vision-Language Interaction Intelligence
dots.tts Technical Report
Deterministic Video Depth Estimation with Generative Priors
Galaxy Image Deconvolution for Weak Gravitational Lensing with Unrolled Plug-and-Play ADMM
AI Must Embrace Specialization via Superhuman Adaptable Intelligence
Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians
Agents of Chaos
HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry
Orchestra-o1: Omnimodal Agent Orchestration
From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents
APPO: Agentic Procedural Policy Optimization
OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data
InterleaveThinker: Reinforcing Agentic Interleaved Generation
MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling
SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning
WEAVEBENCH: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces
MiniMax Sparse Attention
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
Flex4DHuman: Flexible Multi-view Video Diffusion for 4D Human Reconstruction
Modality Forcing for Scalable Spatial Generation
From AGI to ASI
World Tracing: Generative Pixel-Aligned Geometry Beyond the Visible
Regularized f-Divergence Kernel Tests
Pretraining Recurrent Networks without Recurrence
FastContext: Training Efficient Repository Explorer for Coding Agents
VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models
DreamX-World 1.0: A General-Purpose Interactive World Model
Geometric Action Model for Robot Policy Learning
Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories
JoyAI-VL-Interaction: Real-Time Vision-Language Interaction Intelligence
dots.tts Technical Report
Deterministic Video Depth Estimation with Generative Priors
Galaxy Image Deconvolution for Weak Gravitational Lensing with Unrolled Plug-and-Play ADMM
AI Must Embrace Specialization via Superhuman Adaptable Intelligence
Sycophantic Chatbots Cause Delusional Spiraling, Even in Ideal Bayesians
Agents of Chaos
HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry
Orchestra-o1: Omnimodal Agent Orchestration
From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI
Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents
APPO: Agentic Procedural Policy Optimization
OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data
InterleaveThinker: Reinforcing Agentic Interleaved Generation
MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling
SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning
WEAVEBENCH: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces
MiniMax Sparse Attention
EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments
Flex4DHuman: Flexible Multi-view Video Diffusion for 4D Human Reconstruction
Modality Forcing for Scalable Spatial Generation
From AGI to ASI
World Tracing: Generative Pixel-Aligned Geometry Beyond the Visible
Regularized f-Divergence Kernel Tests
Pretraining Recurrent Networks without Recurrence