List your company

Medical Visual Question Answering on Pooled Medical VQA Datasets Open 5-fold CV

0.068ACEIsotonic Regr.
0.048680.179090.30950.43991Apr 2, 2026

Evaluation results

MethodLinks
Isotonic Regr.Backbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.040.068
HAC-PlattBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.040.075
HAC-GateBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.040.075
HAC-GateBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.040.077
Hist. BinningBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.040.079
Platt ScalingBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.040.082
HAC-PlattBackbone=InternVL3-2B, Confidence Type=Sampling2026.040.083
HAC-PlattBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.040.083
HAC-GateBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.040.084
HAC-GateBackbone=InternVL3-2B, Confidence Type=Sampling2026.040.085
HAC-PlattBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.040.086
HAC-GateBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.040.086
HAC-PlattBackbone=InternVL3-38B, Confidence Type=Sampling2026.040.086
Isotonic Regr.Backbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.040.087
Hist. BinningBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.040.087
HAC-PlattBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.040.087
Platt ScalingBackbone=InternVL3-2B, Confidence Type=Sampling2026.040.087
HAC-PlattBackbone=InternVL3-8B, Confidence Type=Verbalized2026.040.087
HAC-GateBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.040.088
Isotonic Regr.Backbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.040.088
HAC-GateBackbone=InternVL3-38B, Confidence Type=Sampling2026.040.088
HAC-PlattBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.040.089
Hist. BinningBackbone=InternVL3-2B, Confidence Type=Sampling2026.040.089
HAC-GateBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.040.09
Isotonic Regr.Backbone=InternVL3-2B, Confidence Type=Sampling2026.040.09
Isotonic Regr.Backbone=InternVL3-8B, Confidence Type=Sampling2026.040.09
HAC-GateBackbone=InternVL3-8B, Confidence Type=Verbalized2026.040.091
Isotonic Regr.Backbone=InternVL3-38B, Confidence Type=Sampling2026.040.091
HAC-GateBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.040.092
Hist. BinningBackbone=InternVL3-38B, Confidence Type=Sampling2026.040.093
Hist. BinningBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.040.093
Isotonic Regr.Backbone=Qwen3-VL-8B, Confidence Type=Sampling2026.040.094
Platt ScalingBackbone=InternVL3-8B, Confidence Type=Sampling2026.040.094
Isotonic Regr.Backbone=InternVL3-8B, Confidence Type=Verbalized2026.040.094
HAC-PlattBackbone=InternVL3-8B, Confidence Type=Sampling2026.040.094
Platt ScalingBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.040.095
Platt ScalingBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.040.095
Platt ScalingBackbone=InternVL3-2B, Confidence Type=Verbalized2026.040.096
Hist. BinningBackbone=InternVL3-8B, Confidence Type=Verbalized2026.040.096
Isotonic Regr.Backbone=InternVL3-38B, Confidence Type=Verbalized2026.040.096
HAC-PlattBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.040.096
HAC-GateBackbone=InternVL3-8B, Confidence Type=Sampling2026.040.097
Platt ScalingBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.040.098
HAC-GateBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.040.098
Platt ScalingBackbone=InternVL3-38B, Confidence Type=Sampling2026.040.099
HAC-PlattBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.040.1
HAC-PlattBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.040.102
Platt ScalingBackbone=InternVL3-8B, Confidence Type=Verbalized2026.040.102
Isotonic Regr.Backbone=Qwen3-VL-32B, Confidence Type=Sampling2026.040.103
Hist. BinningBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.040.104
Isotonic Regr.Backbone=InternVL3-2B, Confidence Type=Verbalized2026.040.104
Hist. BinningBackbone=InternVL3-38B, Confidence Type=Verbalized2026.040.104
Platt ScalingBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.040.104
Hist. BinningBackbone=InternVL3-8B, Confidence Type=Sampling2026.040.105
Platt ScalingBackbone=InternVL3-38B, Confidence Type=Verbalized2026.040.105
Isotonic Regr.Backbone=Qwen3-VL-2B, Confidence Type=Sampling2026.040.106
HAC-GateBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.040.106
Hist. BinningBackbone=InternVL3-2B, Confidence Type=Verbalized2026.040.106
Isotonic Regr.Backbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.040.106
Hist. BinningBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.040.107
Hist. BinningBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.040.107
Platt ScalingBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.040.108
Isotonic Regr.Backbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.040.109
Platt ScalingBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.040.11
Platt ScalingBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.040.111
Hist. BinningBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.040.112
HAC-PlattBackbone=InternVL3-2B, Confidence Type=Verbalized2026.040.115
HAC-PlattBackbone=InternVL3-38B, Confidence Type=Verbalized2026.040.116
HAC-GateBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.040.116
Hist. BinningBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.040.118
HAC-GateBackbone=InternVL3-2B, Confidence Type=Verbalized2026.040.118
Isotonic Regr.Backbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.040.118
HAC-GateBackbone=InternVL3-38B, Confidence Type=Verbalized2026.040.119
Hist. BinningBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.040.119
HAC-PlattBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.040.119
Hist. BinningBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.040.12
Isotonic Regr.Backbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.040.12
HAC-PlattBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.040.12
Platt ScalingBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.040.124
Platt ScalingBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.040.133
UncalibratedBackbone=InternVL3-8B, Confidence Type=Sampling2026.040.228
UncalibratedBackbone=InternVL3-2B, Confidence Type=Sampling2026.040.249
UncalibratedBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.040.256
UncalibratedBackbone=InternVL3-38B, Confidence Type=Sampling2026.040.294
UncalibratedBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.040.321
UncalibratedBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.040.345
UncalibratedBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.040.349
UncalibratedBackbone=InternVL3-8B, Confidence Type=Verbalized2026.040.352
UncalibratedBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.040.36
UncalibratedBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.040.363
UncalibratedBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.040.4
UncalibratedBackbone=InternVL3-38B, Confidence Type=Verbalized2026.040.406
UncalibratedBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.040.411
UncalibratedBackbone=InternVL3-2B, Confidence Type=Verbalized2026.040.421
UncalibratedBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.040.539
UncalibratedBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.040.551