List your company

Visual Question Answering on M3CoT

61.2AccuracyAIM-CoT
24.935234.350143.76553.1799Sep 30, 2025Nov 8, 2025Dec 18, 2025Jan 26, 2026Mar 7, 2026Apr 15, 2026May 25, 2026

Evaluation results

MethodLinks
AIM-CoTBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0961.2
ICoTBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0959.1
AIM-CoTBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0958.7
DDCoTBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0957.5
CCoTBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0956.9
ICoTBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0956.9
No-CoTBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0956.8
SCAFFOLDBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0956.7
CCoTBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0955.3
No-CoTBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0955.2
DDCoTBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0954.8
MMCoTBackbone=Qwen2.5-VL-32B, Shot=1-shot2025.0954.7
SCAFFOLDBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0954.2
MMCoTBackbone=Qwen2.5-VL-32B, Shot=0-shot2025.0953.4
AIM-CoTBackbone=Qwen2-VL-7B, Shot=1-shot2025.0946.6
ICoTBackbone=Qwen2-VL-7B, Shot=1-shot2025.0946
DDCoTBackbone=Qwen2-VL-7B, Shot=1-shot2025.0945.7
No-CoTBackbone=Qwen2-VL-7B, Shot=1-shot2025.0945.4
SCAFFOLDBackbone=Qwen2-VL-7B, Shot=1-shot2025.0944.9
AIM-CoTBackbone=Qwen2-VL-7B, Shot=0-shot2025.0944.7
CCoTBackbone=Qwen2-VL-7B, Shot=1-shot2025.0944.1
ICoTBackbone=Qwen2-VL-7B, Shot=0-shot2025.0944.1
No-CoTBackbone=Qwen2-VL-7B, Shot=0-shot2025.0943.6
CCoTBackbone=Qwen2-VL-7B, Shot=0-shot2025.0943.3
DDCoTBackbone=Qwen2-VL-7B, Shot=0-shot2025.0942.6
MMCoTBackbone=Qwen2-VL-7B, Shot=1-shot2025.0942.5
VRCDdecoding length (L)=384, forward ratio (FR)=0.52026.0541.99
SCAFFOLDBackbone=Qwen2-VL-7B, Shot=0-shot2025.0941.7
AIM-CoTBackbone=Janus-Pro-7B, Shot=1-shot2025.0941.5
MMCoTBackbone=Qwen2-VL-7B, Shot=0-shot2025.0940.1
VRCDdecoding length (L)=384, forward ratio (FR)=0.252026.0539.82
AIM-CoTBackbone=Janus-Pro-7B, Shot=0-shot2025.0939.7
ICoTBackbone=Janus-Pro-7B, Shot=1-shot2025.0939.4
IGdecoding length (L)=384, forward ratio (FR)=0.52026.0539.23
DDCoTBackbone=Janus-Pro-7B, Shot=1-shot2025.0938.6
Margindecoding length (L)=384, forward ratio (FR)=0.52026.0538.47
Confidencedecoding length (L)=384, forward ratio (FR)=0.52026.0538.35
CCoTBackbone=Janus-Pro-7B, Shot=1-shot2025.0938.2
Margindecoding length (L)=384, forward ratio (FR)=0.252026.0538.06
SCAFFOLDBackbone=Janus-Pro-7B, Shot=1-shot2025.0938
No-CoTBackbone=Janus-Pro-7B, Shot=1-shot2025.0937.8
ICoTBackbone=Janus-Pro-7B, Shot=0-shot2025.0937.6
IGdecoding length (L)=384, forward ratio (FR)=0.252026.0537.26
Confidencedecoding length (L)=384, forward ratio (FR)=0.252026.0537.16
DDCoTBackbone=Janus-Pro-7B, Shot=0-shot2025.0936.8
MMCoTBackbone=Janus-Pro-7B, Shot=1-shot2025.0936.6
No-CoTBackbone=Janus-Pro-7B, Shot=0-shot2025.0936.5
CCoTBackbone=Janus-Pro-7B, Shot=0-shot2025.0936.4
SCAFFOLDBackbone=Janus-Pro-7B, Shot=0-shot2025.0936.1
Entropydecoding length (L)=384, forward ratio (FR)=0.52026.0535.86
MMCoTBackbone=Janus-Pro-7B, Shot=0-shot2025.0935.2
Entropydecoding length (L)=384, forward ratio (FR)=0.252026.0534.57
VRCDdecoding length (L)=384, forward ratio (FR)=0.1252026.0534.24
Margindecoding length (L)=384, forward ratio (FR)=0.1252026.0534.11
AIM-CoTBackbone=Chameleon-7B, Shot=1-shot2025.0932.8
ICoTBackbone=Chameleon-7B, Shot=1-shot2025.0932.3
CCoTBackbone=Chameleon-7B, Shot=1-shot2025.0931.4
AIM-CoTBackbone=Chameleon-7B, Shot=0-shot2025.0931.4
SCAFFOLDBackbone=Chameleon-7B, Shot=1-shot2025.0931.1
MMCoTBackbone=Chameleon-7B, Shot=1-shot2025.0930.6
DDCoTBackbone=Chameleon-7B, Shot=1-shot2025.0929.8
ICoTBackbone=Chameleon-7B, Shot=0-shot2025.0929.8
SCAFFOLDBackbone=Chameleon-7B, Shot=0-shot2025.0929.6
CCoTBackbone=Chameleon-7B, Shot=0-shot2025.0929.4
No-CoTBackbone=Chameleon-7B, Shot=0-shot2025.0929.1
Confidencedecoding length (L)=384, forward ratio (FR)=0.1252026.0528.82
IGdecoding length (L)=384, forward ratio (FR)=0.1252026.0528.72
DDCoTBackbone=Chameleon-7B, Shot=0-shot2025.0928.6
MMCoTBackbone=Chameleon-7B, Shot=0-shot2025.0928.5
No-CoTBackbone=Chameleon-7B, Shot=1-shot2025.0928.4
Entropydecoding length (L)=384, forward ratio (FR)=0.1252026.0526.33