TwinPast: Benchmarking Historical Evidence Use in Streaming Video Understanding with Evidence Contrastive Video Pairs

Paper recorded by Signals 4 on 2026-10-30 in cs.CV. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-30 on arXiv · recorded by Signals 4 on 2026-11-02

Category: cs.CV · 计算机视觉 · first seen 2026-11-02

Abstract

High accuracy on streaming video benchmarks does not establish whether a model uses visual history. Our review of OVO-Bench and StreamingBench finds that recent frames suffice for 32% of sampled historical questions, while models also answer many questions correctly from the question and options alone. We introduce TwinPast to test whether answers correctly follow changes in history, using 350 hum

Read on arXiv →