Limin Wang

PixNerd: Pixel Neural Field Diffusion

CaReBench: A Fine-grained Benchmark for Video Captioning and Retrieval

VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling

ExpVid: A Benchmark for Experiment Video Understanding & Reasoning

TokenSculpt: Pruning with Min-Max Spatio-Temporal Duplication for Video Grounding

Balancing the Experts: Unlocking LoRA-MoE for GRPO via Mechanism-Aware Rewards

FreeRet: MLLMs as Training-Free Retrievers

RIVER: A Real-Time Interaction Benchmark for Video LLMs

Learning Goal-Oriented Language-Guided Navigation with Self-Improving Demonstrations at Scale

UniFlow: A Unified Pixel Flow Tokenizer for Visual Understanding and Generation

Benchmarking Visual Knowledge in Multimodal Large Language Models

Arbitrary Generative Video Interpolation

Towards Pixel-level VLM Perception via Simple Points Prediction

VideoCap-R1: Enhancing MLLMs for Video Captioning via Structured Thinking

History-Aware Transformation of ReID Features for Multiple Object Tracking

Efficient Low-rank and Sparse Approximation and Adaptation for Large Language Models

CoMo: Learning Continuous Latent Motion from Internet Videos for Scalable Robot Learning

StreamForest: Efficient Online Video Understanding with Persistent Event Memory

OmniCorpus: A Unified Multimodal Corpus of 10 Billion-Level Images Interleaved with Text

VideoChat-R1.5: Visual Test-Time Scaling to Reinforce Multimodal Reasoning by Iterative Perception

Eagle 2.5: Boosting Long-Context Post-Training for Frontier Vision-Language Models

SPA: 3D Spatial-Awareness Enables Effective Embodied Representation

Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel

MotionRAG: Motion Retrieval-Augmented Image-to-Video Generation

LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization

Differentiable Solver Search for fast diffusion sampling

Differentiable Solver Search for Fast Diffusion Sampling

CG-Bench: Clue-grounded Question Answering Benchmark for Long Video Understanding

Modeling Fine-Grained Hand-Object Dynamics for Egocentric Video Representation Learning

Stochastic Layer-Wise Shuffle for Improving Vision Mamba Training

TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning

RotPruner: Large Language Model Pruning in Rotated Space

Efficient Test-Time Prompt Tuning for Vision-Language Models

DMM: Building a Versatile Image Generation Model via Distillation-Based Model Merging

TrackMamba: Mamba-Transformer Tracking

VideoEval: Comprehensive Benchmark Suite for Low-Cost Evaluation of Video Foundation Model

Tra-MoE: Scaling Trajectory Prediction Models for Adaptive Policy Conditioning

Stochastic Layer-Wise Shuffle: A Good Practice to Improve Vision Mamba Training