TVBench

TVBench is a new benchmark specifically created to evaluate temporal understanding in video QA. We identified three main issues in existing datasets: (i) static information from single frames is often sufficient to solve the tasks (ii) the text of the questions and candidate answers is overly informative, allowing models to answer correctly without relying on any visual input (iii) world knowledge alone can answer many of the questions, making the benchmarks a test of knowledge replication rather than visual reasoning. In addition, we found that open-ended question-answering benchmarks for video understanding suffer from similar issues while the automatic evaluation process with LLMs is unreliable, making it an unsuitable alternative.

We defined 10 temporally challenging tasks that either require repetition counting (Action Count), properties about moving objects (Object Shuffle, Object Count, Moving Direction), temporal localization (Action Localization, Unexpected Action), temporal sequential ordering (Action Sequence, Scene Transition, Egocentric Sequence) and distinguishing between temporally hard Action Antonyms such as "Standing up" and "Sitting down".