Paper-Conference

VMonarch: Efficient Video Diffusion Transformers with Structured Attention

VideoRealBench: A Chain-of-Thought Realism Evaluation Benchmark for Generated Human-Centric Videos

UniAVGen: Unified Audio and Video Generation with Asymmetric Cross-Modal Interactions

TimeLens: Rethinking Video Temporal Grounding with Multimodal LLMs

TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning

Rethinking BCE Loss for Multi-Label Image Recognition with Fine-Tuning

InternVideo-Next: Towards World-Understanding Video Models

DDT: Decoupled Diffusion Transformer

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

Bayesian Decomposition and Semantic Completion for Few-shot Semantic Segmentation