List your company

Medical Visual Question Answering on Pooled Medical VQA Datasets Closed (5-fold CV)

37.1ACEUncalibrated
7.9815.5423.130.66Apr 2, 2026

Evaluation results

MethodLinks
UncalibratedBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.0437.1
UncalibratedBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.0424.1
UncalibratedBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.0422.4
UncalibratedBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.0421.4
UncalibratedBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.0421.2
UncalibratedBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.0421.2
UncalibratedBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.0419.9
UncalibratedBackbone=InternVL3-2B, Confidence Type=Sampling2026.0419.4
UncalibratedBackbone=InternVL3-2B, Confidence Type=Verbalized2026.0419.2
UncalibratedBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.0418
UncalibratedBackbone=InternVL3-8B, Confidence Type=Verbalized2026.0417.6
UncalibratedBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.0416.6
UncalibratedBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.0416.5
UncalibratedBackbone=InternVL3-38B, Confidence Type=Verbalized2026.0416.4
HAC-GateBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.0416.3
UncalibratedBackbone=InternVL3-8B, Confidence Type=Sampling2026.0415.2
Platt ScalingBackbone=InternVL3-2B, Confidence Type=Sampling2026.0414.5
HAC-PlattBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.0414
HAC-PlattBackbone=InternVL3-2B, Confidence Type=Sampling2026.0413.5
Hist. BinningBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.0413.3
Isotonic Regr.Backbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.0413.1
HAC-GateBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.0413.1
Hist. BinningBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.0412.9
HAC-GateBackbone=InternVL3-2B, Confidence Type=Sampling2026.0412.8
Platt ScalingBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.0412.8
HAC-PlattBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.0412.8
Platt ScalingBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.0412.7
HAC-PlattBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.0412.4
HAC-GateBackbone=InternVL3-2B, Confidence Type=Verbalized2026.0412.4
HAC-PlattBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.0412.4
HAC-GateBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.0412.4
Platt ScalingBackbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.0412.3
Isotonic Regr.Backbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.0412.3
HAC-PlattBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.0412.3
UncalibratedBackbone=InternVL3-38B, Confidence Type=Sampling2026.0412.2
HAC-GateBackbone=InternVL3-38B, Confidence Type=Verbalized2026.0412.1
Isotonic Regr.Backbone=InternVL3-2B, Confidence Type=Sampling2026.0411.9
HAC-PlattBackbone=InternVL3-38B, Confidence Type=Verbalized2026.0411.9
HAC-GateBackbone=LLaVA-NeXT-34B, Confidence Type=Sampling2026.0411.9
Hist. BinningBackbone=InternVL3-2B, Confidence Type=Sampling2026.0411.7
Platt ScalingBackbone=InternVL3-38B, Confidence Type=Verbalized2026.0411.7
Platt ScalingBackbone=InternVL3-8B, Confidence Type=Sampling2026.0411.6
Hist. BinningBackbone=LLaVA-NeXT-7B, Confidence Type=Sampling2026.0411.6
HAC-GateBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.0411.5
HAC-GateBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.0411.4
HAC-PlattBackbone=InternVL3-8B, Confidence Type=Sampling2026.0411.4
Hist. BinningBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.0411.3
HAC-GateBackbone=InternVL3-8B, Confidence Type=Sampling2026.0411.3
HAC-GateBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.0411.2
Platt ScalingBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.0411.1
Isotonic Regr.Backbone=InternVL3-2B, Confidence Type=Verbalized2026.0411.1
Isotonic Regr.Backbone=LLaVA-NeXT-7B, Confidence Type=Verbalized2026.0411.1
HAC-PlattBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.0411
Platt ScalingBackbone=InternVL3-2B, Confidence Type=Verbalized2026.0411
Hist. BinningBackbone=InternVL3-8B, Confidence Type=Sampling2026.0411
Isotonic Regr.Backbone=InternVL3-38B, Confidence Type=Verbalized2026.0411
Hist. BinningBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.0411
Platt ScalingBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.0410.9
Isotonic Regr.Backbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.0410.9
Platt ScalingBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.0410.8
HAC-PlattBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.0410.8
HAC-PlattBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.0410.8
Platt ScalingBackbone=LLaVA-NeXT-34B, Confidence Type=Verbalized2026.0410.7
HAC-GateBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.0410.6
Platt ScalingBackbone=Qwen3-VL-2B, Confidence Type=Sampling2026.0410.4
Platt ScalingBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.0410.4
HAC-PlattBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.0410.4
HAC-GateBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.0410.4
Hist. BinningBackbone=InternVL3-38B, Confidence Type=Verbalized2026.0410.4
HAC-PlattBackbone=InternVL3-38B, Confidence Type=Sampling2026.0410.3
Hist. BinningBackbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.0410.2
Isotonic Regr.Backbone=InternVL3-8B, Confidence Type=Sampling2026.0410.2
Platt ScalingBackbone=InternVL3-38B, Confidence Type=Sampling2026.0410.2
HAC-PlattBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.0410.1
HAC-GateBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.0410.1
HAC-PlattBackbone=InternVL3-2B, Confidence Type=Verbalized2026.0410.1
HAC-GateBackbone=InternVL3-38B, Confidence Type=Sampling2026.0410.1
Hist. BinningBackbone=Qwen3-VL-8B, Confidence Type=Sampling2026.0410
Isotonic Regr.Backbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.0410
Isotonic Regr.Backbone=Qwen3-VL-32B, Confidence Type=Sampling2026.049.9
Hist. BinningBackbone=InternVL3-2B, Confidence Type=Verbalized2026.049.9
Isotonic Regr.Backbone=Qwen3-VL-2B, Confidence Type=Sampling2026.049.8
Hist. BinningBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.049.8
Hist. BinningBackbone=InternVL3-8B, Confidence Type=Verbalized2026.049.8
Hist. BinningBackbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.049.7
Hist. BinningBackbone=Qwen3-VL-32B, Confidence Type=Verbalized2026.049.7
Isotonic Regr.Backbone=Qwen3-VL-8B, Confidence Type=Sampling2026.049.6
Platt ScalingBackbone=Qwen3-VL-32B, Confidence Type=Sampling2026.049.6
HAC-PlattBackbone=InternVL3-8B, Confidence Type=Verbalized2026.049.6
Isotonic Regr.Backbone=Qwen3-VL-2B, Confidence Type=Verbalized2026.049.5
Isotonic Regr.Backbone=Qwen3-VL-8B, Confidence Type=Verbalized2026.049.5
HAC-GateBackbone=InternVL3-8B, Confidence Type=Verbalized2026.049.5
Isotonic Regr.Backbone=InternVL3-38B, Confidence Type=Sampling2026.049.5
Platt ScalingBackbone=InternVL3-8B, Confidence Type=Verbalized2026.049.4
Isotonic Regr.Backbone=InternVL3-8B, Confidence Type=Verbalized2026.049.3
Hist. BinningBackbone=InternVL3-38B, Confidence Type=Sampling2026.049.1