Tingting Gao

Pre-indexing hours of video is a computational dead end—growing an online, query-driven textual memory tree during inference matches offline accuracy without the upfront processing tax.

Extensive offline evaluation, human calibration, and online A/B tests show that SARA-7B generates more specific, polarity-consistent, and grounded rationales than strong MLLM baselines, while SARA-Ranker improves engagement and reduces negative feedback in production.

VideoRover unifies video reasoning and external knowledge retrieval, achieving competitive performance with fewer resources than larger models.