AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models

Learning to Reason in 13 Parameters

DFlash: Block Diffusion for Flash Speculative Decoding

Context Forcing: Consistent Autoregressive Video Generation with Long Context

MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

Length-Unbiased Sequence Policy Optimization: Revealing and Controlling Response Length Variation in RLVR

Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening

CAR-bench: Evaluating the Consistency and Limit-Awareness of LLM Agents under Real-World Uncertainty

Streaming Sequence-to-Sequence Learning with Delayed Streams Modeling

Kiss3DGen: Repurposing Image Diffusion Models for 3D Asset Generation

Stateful Conformer with Cache-Based Inference for Streaming Automatic Speech Recognition

Native and Compact Structured Latents for 3D Generation

Continuous Audio Language Models

Evolving Interactive Diagnostic Agents in a Virtual Clinical Environment

WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference

TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times

HunyuanVideo-Foley: Multimodal Diffusion with Representation Alignment for High-Fidelity Foley Audio Generation

Fara-7B: An Efficient Agentic Model for Computer Use

Fun-ASR Technical Report

Accelerating Scientific Research with Gemini: Case Studies and Common Techniques

Scaling Small Agents Through Strategy Auctions

Vibe AIGC: A New Paradigm for Content Generation via Agentic Orchestration

PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR

EgoActor: Grounding Task Planning into Spatial-aware Egocentric Actions for Humanoid Robots via Visual-Language Models

A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces

Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization

SoMA: A Real-to-Sim Neural Simulator for Robotic Soft-body Manipulation

3D-Aware Implicit Motion Control for View-Adaptive Human Video Generation

daVinci-Agency: Unlocking Long-Horizon Agency Data-Efficiently

Research on World Models Is Not Merely Injecting World Knowledge into Specific Tasks

AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models

Learning to Reason in 13 Parameters

DFlash: Block Diffusion for Flash Speculative Decoding

Context Forcing: Consistent Autoregressive Video Generation with Long Context

MemSkill: Learning and Evolving Memory Skills for Self-Evolving Agents

Length-Unbiased Sequence Policy Optimization: Revealing and Controlling Response Length Variation in RLVR

Spider-Sense: Intrinsic Risk Sensing for Efficient Agent Defense with Hierarchical Adaptive Screening

CAR-bench: Evaluating the Consistency and Limit-Awareness of LLM Agents under Real-World Uncertainty

Streaming Sequence-to-Sequence Learning with Delayed Streams Modeling

Kiss3DGen: Repurposing Image Diffusion Models for 3D Asset Generation

Stateful Conformer with Cache-Based Inference for Streaming Automatic Speech Recognition

Native and Compact Structured Latents for 3D Generation

Continuous Audio Language Models

Evolving Interactive Diagnostic Agents in a Virtual Clinical Environment

WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference

TurboDiffusion: Accelerating Video Diffusion Models by 100-200 Times

HunyuanVideo-Foley: Multimodal Diffusion with Representation Alignment for High-Fidelity Foley Audio Generation

Fara-7B: An Efficient Agentic Model for Computer Use

Fun-ASR Technical Report

Accelerating Scientific Research with Gemini: Case Studies and Common Techniques

Scaling Small Agents Through Strategy Auctions

Vibe AIGC: A New Paradigm for Content Generation via Agentic Orchestration

PaperSearchQA: Learning to Search and Reason over Scientific Papers with RLVR

EgoActor: Grounding Task Planning into Spatial-aware Egocentric Actions for Humanoid Robots via Visual-Language Models

A-RAG: Scaling Agentic Retrieval-Augmented Generation via Hierarchical Retrieval Interfaces

Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization

SoMA: A Real-to-Sim Neural Simulator for Robotic Soft-body Manipulation

3D-Aware Implicit Motion Control for View-Adaptive Human Video Generation

daVinci-Agency: Unlocking Long-Horizon Agency Data-Efficiently

Research on World Models Is Not Merely Injecting World Knowledge into Specific Tasks