Yu Wang

SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning

Translate Policy to Language: Flow Matching Generated Rewards for LLM Explanations

MARSHAL: Incentivizing Multi-Agent Reasoning via Self-Play with Strategic LLMs

SAC Flow: Sample-Efficient Reinforcement Learning of Flow-Based Policies via Velocity-Reparameterized Sequential Modeling

Cache-to-Cache: Direct Semantic Communication Between Large Language Models

RE-PO: Robust Enhanced Policy Optimization as a General Framework for LLM Alignment

Think-at-Hard: Selective Latent Iterations to Improve Reasoning Language Models

PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMs

NI Sampling: Accelerating Discrete Diffusion Sampling by Token Order Optimization

FilMaster: Bridging Cinematic Principles and Generative AI for Automated Film Generation

ReXMoE: Reusing Experts with Minimal Overhead in Mixture-of-Experts

Automatic Reward Shaping from Multi-Objective Human Heuristics

Extending Test-Time Scaling: A 3D Perspective with Context, Batch, and Turn

On the Ability of LLMs to Handle Character-Level Perturbations: How Well and How?

VS-Bench: Evaluating VLMs for Strategic Abilities in Multi-Agent Environments

R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing

Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generation

PAROAttention: Pattern-Aware ReOrdering for Efficient Sparse and Quantized Attention in Visual Generation Models

Learning from Suboptimal Data in Continuous Control via Auto-Regressive Soft Q-Network

Distilled Decoding 2: One-step Sampling of Image Auto-regressive Models with Conditional Score Distillation

What Can RL Bring to VLA Generalization? An Empirical Study

Distilled Decoding 1: One-step Sampling of Image Auto-regressive Models with Flow Matching

ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learning

Mixture of Attention Spans: Optimizing LLM Inference Efficiency with Heterogeneous Sliding-Window Lengths

Evaluating LLMs Across Multi-Cognitive Levels: From Medical Knowledge Mastery to Scenario-Based Problem Solving

LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K

Few-shot In-context Preference Learning using Large Language Models

Linear Combination of Saved Checkpoints Makes Consistency and Diffusion Models Better

ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation

Accelerating Auto-regressive Text-to-Image Generation with Training-free Speculative Jacobi Decoding

MoA: Mixture of Sparse Attention for Automatic Large Language Model Compression

LV-Eval: A Balanced Long-Context Benchmark with 5 Length Levels Up to 256K

Efficient Expert Pruning for Sparse Mixture-of-Experts Language Models: Enhancing Performance and Reducing Inference Costs

FlightBench: Benchmarking Learning-based Methods for Ego-vision-based Quadrotors Navigation

VeSX: A Framework Featured by Verification, Self-Correction and In-context Learning for Web Automation Tasks

Towards Accurate and Efficient Sub-8-Bit Integer Training

Reward-Robust RLHF in LLMs

Learning Strategic Language Agents in the Werewolf Game with Iterative Latent Space Policy Optimization