SOURCE-LINKED INTELLIGENCE

Multi-Agent Self-Improving Reinforcement Learning for Video Reasoning

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Video reasoning tasks such as grounded video question answering and temporal grounding require selecting temporal evidence that supports the query. In many current training setups, temporal supervision is applied through local objectives such as boundary regression or span generation, while verification is used mainly to rerank candidate segments at inference time. We study whether a frozen verifier can also guide training. Our multi-agent framework couples a trainable \emph{Grounder} with a frozen \emph{Verifier}: the Grounder samples candidate trajectories and evidence segments, the Verifier

Read original source ↗ Open in workspace

recordType

paper

region

Global