TraveLER: A Modular Multi-LMM Agent Framework for Video Question-Answering

Benchmark Model Rank Results
video-question-answering-on-perception-testTraveLER#5Accuracy (Top-1): 50.2
video-question-answering-on-situatedTraveLER (0-shot)#11Average Accuracy: 44.9
zero-shot-video-question-answer-on-egoschema-1TraveLER#16Accuracy: 53.3
zero-shot-video-question-answer-on-next-qaTraveLER (GPT-4)#11Accuracy: 68.2