Response-G1: Explicit Scene Graph Modeling for Proactive Streaming Video Understanding

TL;DR

Response-G1 introduces an explicit scene graph framework for proactive streaming video understanding, enabling more accurate and interpretable response timing by grounding evidence and queries in shared graph representations.

Contribution

It presents a novel, query-guided scene graph approach that operates in three stages without fine-tuning, improving proactive video response accuracy.

Findings

Outperforms existing methods on benchmark tasks.

Provides more interpretable response decisions.

Effectively models visual evidence with explicit scene graphs.

Abstract

Proactive streaming video understanding requires Video-LLMs to decide when to respond as a video unfolds, a task where existing methods often fall short due to their implicit, query-agnostic modeling of visual evidence. We introduce Response-G1, a novel framework that establishes explicit, structured alignment between the accumulated video evidence and the query's expected response conditions via scene graphs. The framework operates in three fine-tuning-free stages: (1) online query-guided scene graph generation from streaming clips; (2) memory-based retrieval of the most semantically relevant historical scene graphs; and (3) retrieval-augmented trigger prompting for per-frame "silence/response" decisions. By grounding both evidence and conditions in a shared graph representation, Response-G1 achieves more interpretable and accurate response timing decisions. Experimental results on established benchmarks demonstrate the superiority of our method in both proactive and reactive tasks, validating the advantage of explicit scene graph modeling and retrieval in streaming video understanding.