List your company

Medical Visual Question Answering on VQA-Med Closed

0.013ECEInternVL3-38B
-0.00460.11420.2330.3518Apr 2, 2026

Evaluation results

MethodLinks
InternVL3-38BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.013—
InternVL3-38BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.013—
Qwen3-VL-32BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.021—
Qwen3-VL-32BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.021—
Qwen3-VL-32BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.022—
Qwen3-VL-32BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.022—
Qwen3-VL-32BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.022—
InternVL3-8BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.024—
InternVL3-8BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.025—
InternVL3-8BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.026—
InternVL3-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.027—
InternVL3-8BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.028—
Qwen3-VL-8BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.032—
Qwen3-VL-32BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.036—
Qwen3-VL-2BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.037—
LLaVA-NeXT-34BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.039—
InternVL3-38BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.041—
Qwen3-VL-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.042—
Qwen3-VL-2BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.042—
LLaVA-NeXT-34BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.047—
LLaVA-NeXT-7BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.048—
LLaVA-NeXT-34BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.048—
LLaVA-NeXT-34BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.049—
LLaVA-NeXT-34BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.05—
Qwen3-VL-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.051—
LLaVA-NeXT-7BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.051—
Qwen3-VL-2BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.061—
Qwen3-VL-2BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.062—
Qwen3-VL-2BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.062—
Qwen3-VL-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.063—
Qwen3-VL-2BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.064—
Qwen3-VL-8BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.064—
Qwen3-VL-2BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.065—
Qwen3-VL-8BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.065—
LLaVA-NeXT-34BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.065—
InternVL3-38BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.069—
LLaVA-NeXT-7BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.069—
Qwen3-VL-32BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.077—
InternVL3-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.077—
LLaVA-NeXT-7BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.078—
Qwen3-VL-8BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.079—
Qwen3-VL-8BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.081—
Qwen3-VL-2BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.083—
Qwen3-VL-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.085—
Qwen3-VL-8BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.087—
InternVL3-38BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.087—
Qwen3-VL-32BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.088—
InternVL3-8BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.089—
InternVL3-2BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.09—
Qwen3-VL-8BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.091—
InternVL3-2BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.096—
InternVL3-2BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.096—
InternVL3-8BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.101—
InternVL3-8BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.102—
LLaVA-NeXT-7BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.104—
LLaVA-NeXT-7BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.105—
LLaVA-NeXT-34BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.106—
InternVL3-2BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.107—
InternVL3-38BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.109—
Qwen3-VL-32BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.113—
InternVL3-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.113—
InternVL3-38BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.040.113—
InternVL3-2BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.040.114—
LLaVA-NeXT-7BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.040.114—
LLaVA-NeXT-34BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.116—
InternVL3-8BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.128—
Qwen3-VL-8BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.129—
InternVL3-38BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.040.13—
InternVL3-2BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.136—
Qwen3-VL-32BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.040.138—
InternVL3-8BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.146—
InternVL3-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.151—
InternVL3-2BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.154—
Qwen3-VL-8BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.159—
InternVL3-38BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.162—
InternVL3-8BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.163—
LLaVA-NeXT-7BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.169—
InternVL3-2BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.040.178—
InternVL3-2BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.040.178—
InternVL3-38BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.178—
LLaVA-NeXT-7BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.040.181—
LLaVA-NeXT-34BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.040.185—
Qwen3-VL-32BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.2—
InternVL3-2BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.21—
InternVL3-38BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.215—
LLaVA-NeXT-34BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.218—
LLaVA-NeXT-7BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.040.226—
Qwen3-VL-8BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.242—
LLaVA-NeXT-7BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.258—
LLaVA-NeXT-34BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.265—
Qwen3-VL-2BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.286—
LLaVA-NeXT-34BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.291—
Qwen3-VL-32BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.296—
Qwen3-VL-2BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.297—
InternVL3-38BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.040.338—
LLaVA-NeXT-7BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.040.453—
HAC-GateBackbone=Qwen3-VL-2B, Confidence Type=Sampling, Protocol=5-fold CV2026.04—0.152
HAC-GateBackbone=Qwen3-VL-2B, Confidence Type=Verbalized, Protocol=5-fold CV2026.04—0.215
HAC-GateBackbone=Qwen3-VL-8B, Confidence Type=Sampling, Protocol=5-fold CV2026.04—0.2
HAC-GateBackbone=Qwen3-VL-8B, Confidence Type=Verbalized, Protocol=5-fold CV2026.04—0.197