5 papers
TimeThink: Reasoning with Time for Video LLMs
Handong Li, Longteng Guo, Zikang Liu +8
Video reasoning requires models to identify and verify temporally localized evidence within long video sequences. Recent Video Large Language Models (Video-LLMs) have shown promisi…
AdaSpark: Adaptive Sparsity for Efficient Long-Video Understanding
Handong Li, Zikang Liu, Longteng Guo +10
Processing long-form videos with Video Large Language Models (Video-LLMs) is computationally prohibitive. Current efficiency methods often compromise fine-grained perception throug…
Thinking in Streaming Video
Zikang Liu, Longteng Guo, Handong Li +7
Real-time understanding of continuous video streams is essential for interactive assistants and multimodal agents operating in dynamic environments. However, most existing video re…
VRoPE: Rotary Position Embedding for Video Large Language Models
Zikang Liu, Longteng Guo, Yepeng Tang +6
Rotary Position Embedding (RoPE) has shown strong performance in text-based Large Language Models (LLMs), but extending it to video remains a challenge due to the intricate spatiot…
Prefix Grouper: Efficient GRPO Training through Shared-Prefix Forward
Zikang Liu, Tongtian Yue, Yepeng Tang +5
Group Relative Policy Optimization (GRPO) enhances policy learning by computing gradients from relative comparisons among candidate outputs that share a common input prefix. Despit…