2026-10-02 · America/Los_Angeles · 论文 · #1

OneStreamer: Unifying Perception, Memory, and Proactive Response in Streaming Video Interaction

Streaming video LLMs must retain evidence before its relevance to future tasks is known and respond when sufficient evidence becomes available. The challenge is to form reusable factual memory without compromising real-time perception. We introduce OneStreamer, which jointly learns query-independent evidence recording and task response through a shared proactive generation process. Its Proactive Hierarchical Caption Memory (PHCM) produces time-grounded local-detail captions and summaries of completed events. Streaming caption targets supervise the interpretation of observed video prefixes…

热度 63.2 / 100；排名与评分保留该期记录。