List your company

Audio-Visual Question Answering on MUSIC-AVQA (test)

80.7Acc (Avg)VAST
41.49251.67161.8572.029Mar 26, 2022Dec 11, 2022Aug 28, 2023May 14, 2024Jan 29, 2025Oct 16, 2025Jul 4, 2026

Evaluation results

MethodLinks
VAST0-shot=false2023.1280.7———————————————
Crab2025.0378.94—————76.5890.7374.13———————
VALOR0-shot=false2023.1278.9———————————————
TSPM2025.0376.79—————76.9183.6173.51———————
CAE-AVVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT2026.0276.7—————78.283.972.7———————
MoLTVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT, Trainable Params(%)=25.9, Total Params(M)=441.4, Memory (GB)=22025.1176.1—————77.483.172.3———————
QSTar2026.0175.9883.8172.7279.7670.0372.38———85.6472.0580.6383.4684.984.1778.98
CoPL2026.0275.9—————76.877.175.2———————
AVMoEVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT, Trainable Params(%)=42.7, Total Params(M)=456.6, Memory (GB)=11.42025.1175.7—————77.682.771.9———————
AVMoEVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT2026.0275.7—————77.682.771.9———————
MoLTVisual Encoder=Swin-V2-L (shared), Audio Encoder=Swin-V2-L (shared), Trainable Params(%)=19.5, Total Params(M)=366.2, Memory (GB)=1.72025.1174.9—————75.881.371.5———————
DG-SCTVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT, Trainable Params(%)=50, Total Params(M)=520.2, Memory (GB)=11.32025.1174.8—————77.481.970.7———————
DG-SCTVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT2026.0274.8—————77.481.970.7———————
LAVISH2025.0374.46—————75.9780.2271.26———————
AV-Master2025.1074.2283.672.3979.1364.2170.8———87.0267.8579.9586.5586.6186.5878.51
LAVisHVisual Encoder=Swin-V2-L (shared), Audio Encoder=Swin-V2-L (shared), Trainable Params(%)=8.4, Total Params(M)=249.8, Memory (GB)=10.82025.1174—————75.780.470.4———————
LAVisHBackbone=Swin-V2-L (shared)2026.0274—————75.780.470.4———————
CoQoAudio Encoder=Same as AV-Master2025.1073.92——————————78.9——83.777.4
Q-TriM2026.0773.7882.2971.8580.8762.5869.83———85.0565.4977.8484.4687.0285.7677.68
QA-TIGER2026.0173.7483.172.578.5863.9469.59———84.8667.8578.5883.9686.2985.1477.62
QA-TIGER2025.1073.7483.172.578.5863.9469.59———84.8667.8578.5883.9686.2985.1477.62
QA-Tiger2026.0773.7483.172.578.5863.9469.59———84.8667.8578.5883.9686.2985.1477.62
LAVisHVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT, Trainable Params(%)=28.7, Total Params(M)=367.4, Note=results reported in [6]2025.1173.6—————75.479.670.1———————
LAVisHBackbone=Swin-V2-B (shared)2026.0273.6—————75.479.670.1———————
PSTP-Net2025.0373.52—————70.9177.2672.57———————
TSPM2026.0173.5182.1971.8576.2165.7671.17———84.0764.6576.9182.2984.983.6176.79
TSPM2025.1073.5182.1971.8576.2165.7671.17———84.0764.6576.9182.2984.983.6176.79
TSPM2026.0773.5182.1971.8576.2165.7671.17———84.0764.6576.9182.2984.983.6176.79
MettleVisual Encoder=Swin-V2-L, Audio Encoder=HTS-AT, Trainable Params(%)=29.4, Total Params(M)=452.2, Memory (GB)=3.1, Note=reproduced in this paper environment2025.1172.9—————74.97969.4———————
PSOT2026.0172.61——————————78.22——80.0775.29
PSOTAudio Encoder=Same as AV-Master2025.1072.61——————————78.22——80.0775.29
PSTP2026.0172.5776.1871.872.2371.7969———73.9765.5970.9177.1577.3677.2673.52
PSTP-Net2026.0772.5776.1871.872.2371.7969———73.9765.5970.9177.1577.3677.2673.52
COCA2025.0372.33—————75.4275.2369.96———————
AVAF-Net2025.1072.1284.5168.3775.0561.9470.07———83.0969.778.1580.284.4982.3775.9
ST-AVQA2025.0371.59—————73.8774.469.53———————
MAVQA0-shot=false2023.1271.5———————————————
ST-AVQAVisual Encoder=ResNet-18, Audio Encoder=VGG-like, Trainable Params(%)=11.2, Total Params(M)=94.42025.1171.5—————74.17469.5———————
ST-AVQA2026.0271.5—————74.17469.5———————
LAVISH2026.0171.2681.7166.1375.5163.7767.96———82.0965.5675.9778.9881.4380.2274.46
LAVISH2025.1071.2681.7166.1375.5163.7767.96———82.0965.5675.9778.9881.4380.2274.46
LAVISH2026.0771.2681.7166.1375.5163.7767.96———82.0965.5675.9778.9881.4380.2274.46
APL2026.0170.9682.9966.6873.2964.7665.95———82.470.7178.0976.5282.7479.6974.53
APL2025.1070.9682.9966.6873.2964.7665.95———82.470.7178.0976.5282.7479.6974.53
APL2026.0770.9682.9966.6873.2964.7665.95———82.470.7178.0976.5282.7479.6974.53
SHMamba2025.1070.6482.8967.9372.6561.3168.37———82.363.6475.4278.5381.3179.9374.12
COCA2026.0169.9683.569.7266.6364.1265.57———79.3567.6875.4275.175.4375.2372.33
COCA2026.0769.9683.569.7266.6364.1265.57———79.3567.6875.4275.175.4375.2372.33
Spatio-Temporal Grounded Audio-Visual NetworkModel Domain=AVQA2022.0369.5481.8164.5170.866.0163.23——————————
ST-AVQA2025.1069.5481.8164.5170.866.0163.23———78.1867.0574.0671.5676.387471.52
ST-AVQA2026.0769.5481.8164.5170.866.0163.23———78.1867.0574.0671.5676.387471.52
AVST2026.0169.5382.4964.2469.8864.6765.82———77.7867.1773.8773.5275.2774.471.59
QAGL2025.1069.4382.2962.8372.7363.464.36———82.9971.0478.5880.1277.8878.8973.58
Pano-AVQAVisual Encoder=Faster RCNN, Audio Encoder=VGG-like2025.1168.9—————70.772.666.6———————
Pano-AVQA2026.0268.9—————70.772.666.6———————
AVSDVisual Encoder=VGG-19, Audio Encoder=VGG-like2025.1167.4—————68.570.865.5———————
AVSD2026.0267.4—————68.570.865.5———————
MCCD2026.0167.1380.8751.6371.4664.6764.6———83.8771.0479.1479.7876.7378.2472.2
MCCD2026.0767.1380.8771.4651.6364.6764.6———83.8771.0479.1479.7876.7378.2472.2
PanaAVQA2026.0166.9782.0961.365.3863.6762.04———75.7165.9972.1370.5175.7673.1669.53
Pano-AVQAModel Domain=AVQA2022.0366.6481.2159.3364.9164.2263.23——————————
Pano-AVQA2025.1066.6481.2159.3364.9164.2263.23———74.3664.5670.7369.3975.6572.5668.93
Pano-AVQA2026.0766.6481.2159.3364.9164.2263.23———74.3664.5670.7369.3975.6572.5668.93
AVSDModel Domain=AVQA2022.0365.4981.6158.7963.8961.5261.41——————————
AVSD2025.1065.4981.6158.7963.8961.5261.41———72.4161.968.5267.3974.1970.8367.44
AVSD2026.0765.4981.6158.7963.8961.5261.41———72.4161.968.5267.3974.1970.8367.44
AVSD2026.0165.4480.7757.9364.0362.8561.07———72.4762.4668.786674.5370.3167.32
GRUModel Domain=AudioQA2022.0365.1881.7159.4462.6461.8860.07——————————
HME2026.0164.7880.8754.8963.6463.0360.58———73.6563.7469.8967.4270.268.8366.75
HMEModel Domain=VideoQA2022.0364.0580.353.1863.1962.6959.83——————————
HME2025.1064.0580.353.1863.1962.6959.83———74.7663.5670.6167.9769.4668.7666.45
HME2026.0764.0580.353.1863.1962.6959.83———74.7663.5670.6167.9769.4668.7666.45
PSAC2026.0163.679.7655.2261.6661.1359.85———75.0266.84726870.7869.4166.62
PSACModel Domain=VisualQA2022.0363.5277.5955.0263.4261.1759.47——————————
PSAC2025.1063.5277.5955.0263.4261.1759.47———75.6466.0672.0968.6469.7969.2266.54
PSAC2026.0763.5277.5955.0263.4261.1759.47———75.6466.0672.0968.6469.7969.2266.54
CONVLSTMModel Domain=AudioQA2022.0362.2482.9150.8163.0360.2751.58——————————
MCANModel Domain=VisualQA2022.0361.5880.454.4864.9157.2247.57——————————
MCAN2025.1061.5880.454.4864.9157.2247.57———77.555.2469.2571.5670.9371.2465.49
MCAN2026.0761.5880.454.4864.9157.2247.57———77.555.2469.2571.5670.9371.2465.49
MCAN2026.0161.5280.7754.5765.2256.7746.84———78.0757.7470.5871.7671.7671.7665.83
FCNLSTM2026.0160.4282.2946.259.9262.9447.45———70.865.6668.964.5848.0856.2360.81
HCAttnModel Domain=VisualQA2022.0360.1979.149.5159.9755.2556.43——————————
HCAttn2026.0760.1979.149.5159.9755.2556.43———70.2554.9164.5764.0566.3765.2262.3
FCNLSTMModel Domain=AudioQA2022.0360.0682.0146.2859.3462.1547.33——————————
FCNLSTM2026.0760.0682.0146.2859.3462.1547.33———70.4566.2268.8863.8946.7455.2160.34
BiLSTM AttnModel Domain=VisualQA2022.0357.178.3945.8556.9153.0949.76——————————
BiLSTM2026.0757.178.3945.8556.9153.0949.76———70.3547.9262.0564.6464.3364.4859.92
HCRNModel Domain=VideoQA2022.0350.2654.4741.5353.3852.1147.69——————————
HCRN2025.1050.2654.4741.5353.3852.1147.69———68.5950.9262.0564.3961.8163.0855.73
HCRN2026.0750.2654.4741.5353.3852.1147.69———68.5950.9262.0564.3961.8163.0855.73
HCRN2026.0149.8254.1541.7453.2851.0446.72———71.2950.6763.6965.3364.9865.1556.34
OneLLM-7B0-shot=true2023.1247.6———————————————
ChatBridge-13B0-shot=true2023.1243———————————————
AudioZero-shot=true, Method Category=Individual Modalities, Modality=Audio2026.06———————————————27.93
AV-MasterV-Enc.=CLIPL, A-Enc.=VGGish, Z-S=✕, PT=N/A, Params=34M2025.10———————————————78.5
AV-MasterParams=34.07M, FLOPs=6.20G, Latency=89.24ms2025.10———————————————78.51
AV-MasterSize=0.5B2026.05——————79.986.574.2——————78.5
AVAF-NetSize=0.5B2026.05——————78.182.372.1——————75.9
AVicunaV-Enc.=CLIPL, A-Enc.=CLAP, Z-S=✓, PT=1.1M, Params=7B2025.10———————————————49.6