OmniCap-IF: Benchmarking and Improving Instruction Following Abilities for Omni-Video Captioning
Open paper
Benchmark
Model
Rank
Results
audio-visual-question-answering-on-worldsense
OmniCaptioner-IF-7B
#8
Accuracy: 49.4
audio-visual-question-answering-on-dailyomni
OmniCaptioner-IF-7B
#10
Accuracy: 68.4
Rank counts only results with a code link.