Demo-ICL: In-Context Learning for Procedural Video Knowledge Acquisition
Open paper
Benchmark
Model
Rank
Results
video-question-answering-on-video-mmmu
Demo-ICL (7B)
#29
Accuracy: 52.6
video-question-answering-on-mlvu
Demo-ICL (7B)
#35
Accuracy: 70.4
video-question-answering-on-mvbench
Demo-ICL (SFT, 7B)
#36
Avg.: 69.4
Rank counts only results with a code link.