EG-VQA
Emerging5papers using it
2022first seen
EG-VQA is a benchmark that contains 2,067 videos and 11,838 question-answer pairs annotated with supporting temporal evidence, used to evaluate the ability of models to ground their predictions in relevant video evidence through joint reasoning and precise evidence localization.
Papers using EG-VQA (4)
- Learning to Search: A Decision-Based Agent for Knowledge-Based Visual Question AnsweringCC-VQA: Conflict- and Correlation-Aware Method for Mitigating Knowledge Conflict in Knowledge-Based Visual Question AnsweringKnowledge-based Visual Question Answer with Multimodal Processing, Retrieval and FilteringTowards Reasoning-Aware Explainable VQA