MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings

FreeLong++: Training-Free Long Video Generation via Multi-band SpectralFusion

Thinking Beyond Tokens: From Brain-Inspired Intelligence to Cognitive Foundations for Artificial General Intelligence and its Societal Impact

Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning

SciArena: An Open Evaluation Platform for Foundation Models in Scientific Literature Tasks

Holistic Artificial Intelligence in Medicine; improved performance and explainability

Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective

SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning

Listener-Rewarded Thinking in VLMs for Image Preferences

Calligrapher: Freestyle Text Image Customization

VMoBA: Mixture-of-Block Attention for Video Diffusion Models

SMMILE: An Expert-Driven Benchmark for Multimodal Medical In-Context Learning

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

Shape-for-Motion: Precise and Consistent Video Editing with 3D Proxy

From Ideal to Real: Unified and Data-Efficient Dense Prediction for Real-World Scenarios

ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models

XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation

Zero-shot antibody design in a 24-well plate

KinFormer: Generalizable Dynamical Symbolic Regression for Catalytic Organic Reaction Kinetics

MiCo: Multi-image Contrast for Reinforcement Visual Reasoning

Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs

Ark: An Open-source Python-based Framework for Robot Learning

Pangu Pro MoE: Mixture of Grouped Experts for Efficient Sparsity

LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs

BlenderFusion: 3D-Grounded Visual Editing and Generative Compositing

UniMate: A Unified Model for Mechanical Metamaterial Generation, Property Prediction, and Condition Confirmation

Learning to Skip the Middle Layers of Transformers

SAM4D: Segment Anything in Camera and LiDAR Streams

Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test

FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language

MoCa: Modality-aware Continual Pre-training Makes Better Bidirectional Multimodal Embeddings

FreeLong++: Training-Free Long Video Generation via Multi-band SpectralFusion

Thinking Beyond Tokens: From Brain-Inspired Intelligence to Cognitive Foundations for Artificial General Intelligence and its Societal Impact

Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning

SciArena: An Open Evaluation Platform for Foundation Models in Scientific Literature Tasks

Holistic Artificial Intelligence in Medicine; improved performance and explainability

Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective

SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning

Listener-Rewarded Thinking in VLMs for Image Preferences

Calligrapher: Freestyle Text Image Customization

VMoBA: Mixture-of-Block Attention for Video Diffusion Models

SMMILE: An Expert-Driven Benchmark for Multimodal Medical In-Context Learning

The Automated LLM Speedrunning Benchmark: Reproducing NanoGPT Improvements

Shape-for-Motion: Precise and Consistent Video Editing with 3D Proxy

From Ideal to Real: Unified and Data-Efficient Dense Prediction for Real-World Scenarios

ShotBench: Expert-Level Cinematic Understanding in Vision-Language Models

XVerse: Consistent Multi-Subject Control of Identity and Semantic Attributes via DiT Modulation

Zero-shot antibody design in a 24-well plate

KinFormer: Generalizable Dynamical Symbolic Regression for Catalytic Organic Reaction Kinetics

MiCo: Multi-image Contrast for Reinforcement Visual Reasoning

Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs

Ark: An Open-source Python-based Framework for Robot Learning

Pangu Pro MoE: Mixture of Grouped Experts for Efficient Sparsity

LLaVA-Scissor: Token Compression with Semantic Connected Components for Video LLMs

BlenderFusion: 3D-Grounded Visual Editing and Generative Compositing

UniMate: A Unified Model for Mechanical Metamaterial Generation, Property Prediction, and Condition Confirmation

Learning to Skip the Middle Layers of Transformers

SAM4D: Segment Anything in Camera and LiDAR Streams

Where to find Grokking in LLM Pretraining? Monitor Memorization-to-Generalization without Test

FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language