List your company

Text-to-audio retrieval on AudioCaps

55.2Recall@1InternVideo2-6B
4.65617.77830.944.022Mar 22, 2024Jul 27, 2024Dec 1, 2024Apr 8, 2025Aug 13, 2025Dec 18, 2025Apr 25, 2026

Evaluation results

MethodLinks
InternVideo2-6BEvaluation protocol=Finetuning2024.0355.2———
VASTEvaluation protocol=Finetuning2024.0352———
PEAV LA-Enc Params.=1.1B, Data (M)=124, Sampling=30 FPS2025.1245.8———
PEAV LA-Enc Params.=1.1B, Data (M)=124, Sampling=16 Frames2025.1244.7———
PEAV L-OODA-Enc Params.=1.1B, Data (M)=114, Sampling=30 FPS, OOD=true2025.1243.4———
PEAV BA-Enc Params.=.2B, Data (M)=124, Sampling=16 Frames2025.1243.1———
PEAV BA-Enc Params.=0.2B, Data (M)=124, Sampling=30 FPS2025.1242.7———
PEAV SA-Enc Params.=.09B, Data (M)=124, Sampling=30 FPS2025.1241.8———
PEAV SA-Enc Params.=.09B, Data (M)=124, Sampling=16 Frames2025.1241.2———
MTC (AudioSet)Label=Tag, Evaluation Protocol=audio-language alignment2026.0340.46———
InternVideo2-6BEvaluation protocol=Zero-shot2024.0337.1———
PMRLSetting=Zero-shot2025.0736.175.9——
CLAP-FusionA-Enc Params.=.03B, Data (M)=32025.1235.4———
ProposedModality=t2a, audio encoder=HTSAT-tiny, text encoder=RoBERTa-large2026.0435.284.471.552.1
Proposed audio-text retrieval framework2026.0435.284.471.552.1
GRAMSetting=Zero-shot2025.0734.677.4——
LAION-CLAPModality=t2a2026.0434.182.769.749
PEAV L (PT)A-Enc Params.=1.1B, Data (M)=92, Sampling=30 FPS, PT=true2025.1233.7———
VASTSetting=Zero-shot2025.0733.777.1——
CLAPA-Enc Params.=.03B, Data (M)=32025.1231.6———
AFlamingo2A-Enc Params.=0.3B, Data (M)=82025.1229.8———
Contrastive-scratch (Ours)Label=Caption*, Evaluation Protocol=audio-language alignment2026.0329.66———
Microsoft-CLAPModality=t2a2026.0428.9776343.2
Contrastive-scratchLabel=Caption, Evaluation Protocol=audio-language alignment2026.0328.73———
VIP-ANTEvaluation protocol=Zero-shot2024.0327.7———
VIP-ANTSetting=Zero-shot2025.0727.737.7——
M2D-CLAPA-Enc Params.=.09B, Data (M)=22025.1227.4———
Multi-Task (Ours)Label=Tag*, Caption*, Evaluation Protocol=audio-language alignment2026.0327.07———
PAR (Ours-UTS)Label=Tag Sequence*, Evaluation Protocol=audio-language alignment2026.0326.76———
Captioning-scratch (Ours)Label=Caption*, Evaluation Protocol=audio-language alignment2026.0326.24———
MTC (Ours-UTS)Label=Tag*, Evaluation Protocol=audio-language alignment2026.0326.14———
Captioning-scratchLabel=Caption, Evaluation Protocol=audio-language alignment2026.0326.03———
GeminiRetrieval=BM252026.0425.67360.140.2
AudioGPTRetrieval=BM252026.0425.572.760.941.7
MS-CLAP23’A-Enc Params.=.08B, Data (M)=0.12025.1223.4———
GeminiRetrieval=Semantic2026.0421.850.448.131.7
LTURetrieval=BM252026.0421.160.357.330.8
PengiRetrieval=BM252026.0420.953.250.323.9
GAMARetrieval=BM252026.0420.859.75229.4
LangBindA-Enc Params.=0.3B, Data (M)=102025.1219.7———
LanguageBindSetting=Zero-shot2025.0719.767.6——
SalmonnRetrieval=BM252026.0419.559.648.122.8
GeminiRetrieval=Lexical2026.0419.342.839.324.1
LTURetrieval=Semantic2026.0418.946.739.521.8
AudioGPTRetrieval=Semantic2026.0418.448.547.228.8
SalmonnRetrieval=Semantic2026.0417.947.238.620
SalmonnRetrieval=Lexical2026.0417.639.229.414.7
PengiRetrieval=Semantic2026.0417.339.935.218.6
LTURetrieval=Lexical2026.0417.242.230.320.1
AudioGPTRetrieval=Lexical2026.0416.539.238.721.9
PengiRetrieval=Lexical2026.0412.836.230.718.4
GAMARetrieval=Semantic2026.0412.840.839.319.9
AVFICSetting=Zero-shot, Variant=22025.0710.645.2——
GAMARetrieval=Lexical2026.0410.533.831.415.8
ImageBindSetting=Zero-shot2025.079.342.3——
AVFICSetting=Zero-shot, Variant=12025.078.737.7——
ImageBindA-Enc Params.=.09B, Data (M)=32025.126.6———