Rethink-ACT (AST + TF + MIL)

Reported on 6 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio6 results

Audio captioningonAudioCaps
BLEU-4
0.285
best: 14.3 (Audio Flamingo)
Audio captioningonAudioCaps
CIDEr
0.764
best: 50.2 (Audio Flamingo)
Audio captioningonAudioCaps
METEOR
0.242
best: 20.5 (Audio Flamingo)
Audio captioningonAudioCaps
ROUGE-L
0.504
best: 40.8 (Audio Flamingo)
Audio captioningonAudioCaps
SPICE
0.18
best: 15.1 (Audio Flamingo)
Audio captioningonAudioCaps
SPIDEr
0.472
best: 32.6 (Audio Flamingo)