Sound Probing on Sound Probing v1 (test)
4A@1Contrastive Probe
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Contrastive ProbeAudio Model=PaSST, Language Model=GPT-2 XL, Embedding Type=Language Embedding2024.02 | 4 | 25 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=LLaMA-7B, Embedding Type=Language Embedding2024.02 | 4 | 24 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=BERT Large, Embedding Type=Language Embedding2024.02 | 3 | 20 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=GloVe CC-840B, Embedding Type=Language Embedding2024.02 | 3 | 21 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=BERT, Embedding Type=Language Embedding2024.02 | 2 | 7 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=BERT Large, Embedding Type=Language Embedding2024.02 | 2 | 8 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=GPT-2 XL, Embedding Type=Language Embedding2024.02 | 2 | 10 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=T5, Embedding Type=Language Embedding2024.02 | 2 | 7 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=T5 Large, Embedding Type=Language Embedding2024.02 | 2 | 8 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=LLaMA-7B, Embedding Type=Language Embedding2024.02 | 2 | 9 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=GloVe CC-42B, Embedding Type=Language Embedding2024.02 | 2 | 8 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=GloVe CC-840B, Embedding Type=Language Embedding2024.02 | 2 | 7 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=BERT, Embedding Type=Language Embedding2024.02 | 2 | 11 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=BERT Large, Embedding Type=Language Embedding2024.02 | 2 | 12 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=GPT-2 XL, Embedding Type=Language Embedding2024.02 | 2 | 13 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=T5 Large, Embedding Type=Language Embedding2024.02 | 2 | 11 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=LLaMA-7B, Embedding Type=Language Embedding2024.02 | 2 | 13 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=GloVe CC-42B, Embedding Type=Language Embedding2024.02 | 2 | 10 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=GloVe CC-840B, Embedding Type=Language Embedding2024.02 | 2 | 10 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=BERT, Embedding Type=Language Embedding2024.02 | 2 | 16 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=GPT-2, Embedding Type=Language Embedding2024.02 | 2 | 15 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=T5, Embedding Type=Language Embedding2024.02 | 2 | 20 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=T5 Large, Embedding Type=Language Embedding2024.02 | 2 | 21 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=GloVe CC-42B, Embedding Type=Language Embedding2024.02 | 2 | 20 | |
| Contrastive ProbeAudio Model=PaSST, Language Model=word2vec GNews-300, Embedding Type=Language Embedding2024.02 | 2 | 17 | |
| Contrastive ProbeAudio Model=PANN, Language Model=BERT, Embedding Type=Language Embedding2024.02 | 2 | 5 | |
| Contrastive ProbeAudio Model=PANN, Language Model=GPT-2, Embedding Type=Language Embedding2024.02 | 2 | 6 | |
| Contrastive ProbeAudio Model=PANN, Language Model=GPT-2 XL, Embedding Type=Language Embedding2024.02 | 2 | 8 | |
| Contrastive ProbeAudio Model=PANN, Language Model=T5 Large, Embedding Type=Language Embedding2024.02 | 2 | 5 | |
| Contrastive ProbeAudio Model=PANN, Language Model=LLaMA-7B, Embedding Type=Language Embedding2024.02 | 2 | 6 | |
| Contrastive ProbeAudio Model=PANN, Language Model=GloVe CC-840B, Embedding Type=Language Embedding2024.02 | 2 | 5 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=GPT-2, Embedding Type=Language Embedding2024.02 | 1 | 6 | |
| Contrastive ProbeAudio Model=AudioMAE, Language Model=word2vec GNews-300, Embedding Type=Language Embedding2024.02 | 1 | 6 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=GPT-2, Embedding Type=Language Embedding2024.02 | 1 | 8 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=T5, Embedding Type=Language Embedding2024.02 | 1 | 10 | |
| Contrastive ProbeAudio Model=AudioMAE-FT, Language Model=word2vec GNews-300, Embedding Type=Language Embedding2024.02 | 1 | 8 | |
| Contrastive ProbeAudio Model=PANN, Language Model=BERT Large, Embedding Type=Language Embedding2024.02 | 1 | 4 | |
| Contrastive ProbeAudio Model=PANN, Language Model=T5, Embedding Type=Language Embedding2024.02 | 1 | 4 | |
| Contrastive ProbeAudio Model=PANN, Language Model=GloVe CC-42B, Embedding Type=Language Embedding2024.02 | 1 | 4 | |
| Contrastive ProbeAudio Model=PANN, Language Model=word2vec GNews-300, Embedding Type=Language Embedding2024.02 | 1 | 3 |