List your company

Embodied Task Planning on VoTa-Bench (Seen)

84.72SR (Examine&Light)D2PO
-2.17220.386542.94565.5035Mar 13, 2025

Evaluation results

MethodLinks
D2POBackbone=Qwen2-VL (7B)2025.0384.7266.6784.5271.2712.510.2348.2148.2166.3666.3644.4444.3358.1153.33
DPOBackbone=LLaMA-3.2-Vision (11B)2025.0376.3959.3178.5762.6112.59.9729.4625.4743.9343.3536.5134.2446.0839.73
D2POBackbone=LLaMA-3.2-Vision (11B)2025.0376.3959.6388.171.3214.5812.1938.3932.9748.648.2639.6838.851.1844.84
DPOBackbone=Qwen2-VL (7B)2025.0372.2256.6780.9566.310.428.4744.6444.6460.7560.7544.4444.0453.9249.37
SFTBackbone=Qwen2-VL (7B)2025.0370.8355.2469.0557.746.255.3826.7926.0458.8858.3431.7531.1144.6340.33
D2POBackbone=LLaVA-1.6 (7B)2025.0369.4452.678.5765.4822.9219.647.3247.3260.7560.4144.4444.3354.8350.23
DPOBackbone=LLaVA-1.6 (7B)2025.0366.6745.7772.6259.1720.8318.244.6444.6444.8644.8643.6543.0749.5444.38
SFTBackbone=LLaMA-3.2-Vision (11B)2025.0358.3344.1372.6247.048.336.6930.3626.0346.7346.7335.7131.9842.9935.33
SFTBackbone=LLaVA-1.6 (7B)2025.0356.9445.3763.151.6512.59.8131.2531.1850.4750.0830.1629.3441.3537.56
GPT-4oProtocol=ICL2025.0341.6730.664.2945.954.171.311.791.7924.323.8111.911.3923.518.78
Gemini-1.5-proProtocol=Zero-shot2025.0334.7229.3827.3812.0700007.487.373.171.7210.936.81
Qwen2-VLScale=72B, Protocol=Zero-shot2025.0334.7221.6239.2921.8100003.973.470.790.5611.667.1
GPT-4oProtocol=Zero-shot2025.0333.3323.3751.1936.2700008.416.552.382.0214.3910.37
Qwen2-VLScale=7B, Protocol=Zero-shot2025.0326.398.5514.298.222.080.60000005.832.46
Qwen2-VLScale=7B, Protocol=ICL2025.03259.2521.4312.29000000006.563.14
GPT-4o-miniProtocol=Zero-shot2025.0322.2210.8814.298.16000000005.12.68
LLaVA-1.6Scale=34B, Protocol=Zero-shot2025.0312.52.097.142.67000000002.730.68
LLaMA-3.2-VisionScale=11B, Protocol=Zero-shot2025.0312.524.760.86000000002.370.39
LLaMA-3.2-VisionScale=11B, Protocol=ICL2025.038.331.333.570.57000000001.640.26
LLaVA-1.6Scale=7B, Protocol=ICL2025.031.390.224.760.76000000000.910.15
LLaVA-1.6Scale=7B, Protocol=Zero-shot2025.031.170.677.141.14000000001.640.26