Visual Question Answering on PhD-icc
64.6AccuracyCALRD
Evaluation Results
| Method | Links | |
|---|---|---|
| CALRDDecoding Strategy=Beam, Model=Qwen3-VL2026.06 | 64.6 | |
| CALRDDecoding Strategy=Nucleus, Model=Qwen3-VL2026.06 | 64.52 | |
| CALRDDecoding Strategy=Greedy, Model=Qwen3-VL2026.06 | 63.68 | |
| VCDDecoding Strategy=Nucleus, Model=Qwen3-VL2026.06 | 63.4 | |
| OPERADecoding Strategy=Beam, Model=Qwen3-VL2026.06 | 62.62 | |
| VanillaDecoding Strategy=Beam, Model=Qwen3-VL2026.06 | 62.53 | |
| DeCoDecoding Strategy=Greedy, Model=Qwen3-VL2026.06 | 62.25 | |
| VanillaDecoding Strategy=Nucleus, Model=Qwen3-VL2026.06 | 62.23 | |
| DeCoDecoding Strategy=Nucleus, Model=Qwen3-VL2026.06 | 61.5 | |
| DeCoDecoding Strategy=Beam, Model=Qwen3-VL2026.06 | 61.25 | |
| VanillaDecoding Strategy=Greedy, Model=Qwen3-VL2026.06 | 60.32 | |
| DoLaDecoding Strategy=Greedy, Model=Qwen3-VL2026.06 | 58 | |
| CALRDDecoding Strategy=Beam, Model=Qwen2.5-VL2026.06 | 56.7 | |
| CALRDDecoding Strategy=Greedy, Model=Qwen2.5-VL2026.06 | 56.2 | |
| CALRDDecoding Strategy=Nucleus, Model=Qwen2.5-VL2026.06 | 55.83 | |
| VCDDecoding Strategy=Nucleus, Model=Qwen2.5-VL2026.06 | 53.83 | |
| OPERADecoding Strategy=Beam, Model=Qwen2.5-VL2026.06 | 53.7 | |
| DeCoDecoding Strategy=Greedy, Model=Qwen2.5-VL2026.06 | 51.65 | |
| DeCoDecoding Strategy=Nucleus, Model=Qwen2.5-VL2026.06 | 51.62 | |
| VanillaDecoding Strategy=Greedy, Model=Qwen2.5-VL2026.06 | 51.3 | |
| DeCoDecoding Strategy=Beam, Model=Qwen2.5-VL2026.06 | 51.13 | |
| VanillaDecoding Strategy=Beam, Model=Qwen2.5-VL2026.06 | 51.11 | |
| DoLaDecoding Strategy=Greedy, Model=Qwen2.5-VL2026.06 | 51.08 | |
| VanillaDecoding Strategy=Nucleus, Model=Qwen2.5-VL2026.06 | 51.05 | |
| CALRDDecoding Strategy=Beam, Model=InstructBLIP2026.06 | 49.75 | |
| CALRDDecoding Strategy=Greedy, Model=InstructBLIP2026.06 | 49.62 | |
| CALRDDecoding Strategy=Nucleus, Model=InstructBLIP2026.06 | 48.62 | |
| DeCoDecoding Strategy=Beam, Model=InstructBLIP2026.06 | 42.15 | |
| DeCoDecoding Strategy=Greedy, Model=InstructBLIP2026.06 | 42.12 | |
| DeCoDecoding Strategy=Nucleus, Model=InstructBLIP2026.06 | 41.83 | |
| VanillaDecoding Strategy=Greedy, Model=InstructBLIP2026.06 | 41.08 | |
| VanillaDecoding Strategy=Beam, Model=InstructBLIP2026.06 | 40.73 | |
| VCDDecoding Strategy=Nucleus, Model=InstructBLIP2026.06 | 40.01 | |
| OPERADecoding Strategy=Beam, Model=InstructBLIP2026.06 | 39.78 | |
| VanillaDecoding Strategy=Nucleus, Model=InstructBLIP2026.06 | 39.23 | |
| CALRDDecoding Strategy=Beam, Model=LLaVA-1.62026.06 | 37.4 | |
| CALRDDecoding Strategy=Beam, Model=LLaVA-1.52026.06 | 36.58 | |
| CALRDDecoding Strategy=Greedy, Model=LLaVA-1.62026.06 | 36.52 | |
| CALRDDecoding Strategy=Greedy, Model=LLaVA-1.52026.06 | 36.35 | |
| CALRDDecoding Strategy=Nucleus, Model=LLaVA-1.52026.06 | 34.95 | |
| DeCoDecoding Strategy=Beam, Model=LLaVA-1.62026.06 | 34.63 | |
| CALRDDecoding Strategy=Nucleus, Model=LLaVA-1.62026.06 | 34.12 | |
| DeCoDecoding Strategy=Greedy, Model=LLaVA-1.62026.06 | 33.65 | |
| DeCoDecoding Strategy=Nucleus, Model=LLaVA-1.62026.06 | 32.9 | |
| DoLaDecoding Strategy=Greedy, Model=InstructBLIP2026.06 | 32.65 | |
| VCDDecoding Strategy=Nucleus, Model=LLaVA-1.52026.06 | 31.87 | |
| VCDDecoding Strategy=Nucleus, Model=LLaVA-1.62026.06 | 31.7 | |
| DeCoDecoding Strategy=Nucleus, Model=LLaVA-1.52026.06 | 31.25 | |
| DeCoDecoding Strategy=Beam, Model=LLaVA-1.52026.06 | 31.07 | |
| DeCoDecoding Strategy=Greedy, Model=LLaVA-1.52026.06 | 31 | |
| VanillaDecoding Strategy=Nucleus, Model=LLaVA-1.62026.06 | 30.08 | |
| OPERADecoding Strategy=Beam, Model=LLaVA-1.62026.06 | 29.95 | |
| DoLaDecoding Strategy=Greedy, Model=LLaVA-1.52026.06 | 29.78 | |
| VanillaDecoding Strategy=Nucleus, Model=LLaVA-1.52026.06 | 29.33 | |
| OPERADecoding Strategy=Beam, Model=LLaVA-1.52026.06 | 28.93 | |
| VanillaDecoding Strategy=Beam, Model=LLaVA-1.62026.06 | 28.8 | |
| DoLaDecoding Strategy=Greedy, Model=LLaVA-1.62026.06 | 28.73 | |
| VanillaDecoding Strategy=Beam, Model=LLaVA-1.52026.06 | 28.25 | |
| VanillaDecoding Strategy=Greedy, Model=LLaVA-1.62026.06 | 28.23 | |
| VanillaDecoding Strategy=Greedy, Model=LLaVA-1.52026.06 | 27.72 |