activity
20242026
collaborators

27 papers

cs.CV2026

JoyAI-Video-Edit: Real-Time Open-Ended Video Editing with Autoregressive Diffusion

Yicheng Xiao, Wenxun Dai, Xinran Qin +22

Real-time video editing requires low-latency causal generation with bounded computational resources while preserving source fidelity and long-term temporal consistency. We present…

cs.CV2026

Reinforcing Dual-Path Reasoning in Spatial Vision Language Models

Yatai Ji, An-Chieh Cheng, Yang Fu +13

Spatial VLMs have made substantial progress in geometric perception, yet complex spatial reasoning requiring multi-step inference over depth, distance, and scene relations remains…

cs.LG2026

Smaller Models are Natural Explorers for Policy-Level Diversity in GRPO

Yiming Ren, Yiran Xu, Zicheng Lin +8

We identify a new dimension for enhancing rollout diversity in Group Relative Policy Optimization (GRPO) for LLMs. While GRPO relies on diverse rollouts, prevailing strategies prim…

cs.CV2026

LongLive-RAG: A General Retrieval-Augmented Framework for Long Video Generation

Qixin Hu, Shuai Yang, Wei Huang +2

Autoregressive (AR) video diffusion enables variable-length synthesis, but long-horizon generation often suffers from accumulated errors and identity drift. For efficiency, existin…

cs.CV2026

StreamingVLM: Real-Time Understanding for Infinite Video Streams

Ruyi Xu, Guangxuan Xiao, Yukang Chen +3

Vision-language models (VLMs) could power real-time assistants and autonomous agents, but they face a critical challenge: understanding near-infinite video streams without escalati…

cs.CV2026

LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation

Yukang Chen, Luozhou Wang, Wei Huang +13

We present LongLive-2.0, an NVFP4-based parallel infrastructure throughout the full training and inference workflow of long video generation, addressing speed and memory bottleneck…