EG-VQA: Benchmarking Verifiable Video Question Answering with Grounded Temporal Evidence
For researchers and developers of Video-LLMs, this work exposes a critical gap between answer accuracy and evidence grounding, and provides a benchmark and method to address it.
EG-VQA introduces a benchmark for video question answering that requires models to both answer questions and ground their answers in temporal video evidence. The proposed EG-Reasoner model, trained with explicit evidence supervision, achieves state-of-the-art performance among open-source models and competitive results against proprietary systems, with notable gains on reasoning-intensive tasks like counterfactual questions.
Recent advances in Video Large Language Models (Video-LLMs) have yielded promising performance on video question answering (VideoQA). Nevertheless, existing benchmarks are predominantly evaluated through answer correctness, while the grounding of predictions in relevant video evidence remains largely unexamined. This disconnect between answer generation and evidence understanding motivates the construction of the Evidence-Grounded Video Question Answering Benchmark (EG-VQA), an open-ended evaluation protocol in which each QA pair is explicitly annotated with supporting temporal evidence, thereby requiring joint reasoning and precise evidence localization. EG-VQA is comprised of 2,067 videos and 11,838 QA pairs with fine-grained evidence annotations. To evaluate predicted evidence, Evidence-Grounded F1 (EG-F1) is introduced as a unified metric in which temporal alignment and semantic consistency against ground-truth evidence are jointly measured. Experimental evaluation reveals that even strong proprietary models struggle to accurately ground their predictions, exposing a fundamental discrepancy between answer correctness and faithful evidence localization. To bridge this gap, EG-Reasoner, an evidence-grounded reasoning model trained with explicit supervision, is proposed. State-of-the-art performance is achieved among open-source models, with results competitive against proprietary systems, particularly pronounced gains are observed on reasoning-intensive tasks such as counterfactual questions. These findings demonstrate that scaling alone is insufficient for robust video understanding and that structured evidence supervision is essential for the development of more reliable and interpretable VideoQA systems.