List your company

Audio-to-Text Retrieval on AudioCaps

45.1R@1Proposed
4.33214.91625.536.084Apr 25, 2026

Evaluation results

MethodLinks
ProposedModality=a2t, audio encoder=HTSAT-tiny, text encoder=RoBERTa-large2026.0445.179.390.548.6
Proposed audio-text retrieval framework2026.0445.179.390.548.6
LAION-CLAPModality=a2t2026.0444.476.888.943.8
Microsoft-CLAPModality=a2t2026.0438.169.781.431.9
AudioGPTRetrieval=BM252026.0428.744.175.114.3
GeminiRetrieval=BM252026.0426.648.576.213.6
AudioGPTRetrieval=Semantic2026.0421.637.362.310.4
GeminiRetrieval=Semantic2026.0421.438.542.817.2
GeminiRetrieval=Lexical2026.0420.228.938.815.3
SalmonnRetrieval=BM252026.0419.528.738.220.9
AudioGPTRetrieval=Lexical2026.0418.329.557.29.3
SalmonnRetrieval=Semantic2026.0417.221.630.319.8
SalmonnRetrieval=Lexical2026.0412.520.12213.8
LTURetrieval=BM252026.0411.731.540.311.8
GAMARetrieval=BM252026.0411.229.239.410.4
LTURetrieval=Semantic2026.0410.128.937.29.7
PengiRetrieval=BM252026.041020.734.210.2
GAMARetrieval=Semantic2026.049.318.423.89.3
LTURetrieval=Lexical2026.049.217.328.88.4
PengiRetrieval=Semantic2026.048.118.627.59.5
GAMARetrieval=Lexical2026.047.315.320.67.2
PengiRetrieval=Lexical2026.045.914.720.43.7