List your company

Medical Visual Question Answering on SLAKE Open (ACE)

45.2ACELLaVA-NeXT-34B
8.59218.09627.637.104Apr 2, 2026

Evaluation results

MethodLinks
LLaVA-NeXT-34BCalibration=Uncalibrated, Confidence=Verbalized2026.0445.2
LLaVA-NeXT-7BCalibration=Uncalibrated, Confidence=Verbalized2026.0444.6
InternVL3-38BCalibration=Uncalibrated, Confidence=Verbalized2026.0435.1
InternVL3-2BCalibration=Uncalibrated, Confidence=Verbalized2026.0433.9
Qwen3-VL-2BCalibration=Uncalibrated, Confidence=Verbalized2026.0433.7
Qwen3-VL-8BCalibration=Uncalibrated, Confidence=Sampling2026.0433.1
Qwen3-VL-8BCalibration=Uncalibrated, Confidence=Verbalized2026.0429.9
LLaVA-NeXT-34BCalibration=Uncalibrated, Confidence=Sampling2026.0429.9
Qwen3-VL-2BCalibration=Uncalibrated, Confidence=Sampling2026.0428
Qwen3-VL-32BCalibration=Uncalibrated, Confidence=Sampling2026.0426.5
InternVL3-8BCalibration=Uncalibrated, Confidence=Verbalized2026.0426.5
Qwen3-VL-32BCalibration=Uncalibrated, Confidence=Verbalized2026.0425.9
LLaVA-NeXT-7BCalibration=Uncalibrated, Confidence=Sampling2026.0425
LLaVA-NeXT-34BCalibration=Hist. Binning, Confidence=Sampling2026.0423.9
InternVL3-2BCalibration=Uncalibrated, Confidence=Sampling2026.0423.2
InternVL3-38BCalibration=Uncalibrated, Confidence=Sampling2026.0422.8
LLaVA-NeXT-34BCalibration=Isotonic Regr., Confidence=Sampling2026.0422.7
LLaVA-NeXT-34BCalibration=Platt Scaling, Confidence=Sampling2026.0422.3
InternVL3-8BCalibration=Uncalibrated, Confidence=Sampling2026.0422
LLaVA-NeXT-34BCalibration=HAC-Gate, Confidence=Sampling2026.0420
LLaVA-NeXT-34BCalibration=Isotonic Regr., Confidence=Verbalized2026.0419.8
LLaVA-NeXT-34BCalibration=HAC-Platt, Confidence=Sampling2026.0419.7
LLaVA-NeXT-34BCalibration=Platt Scaling, Confidence=Verbalized2026.0419.6
LLaVA-NeXT-34BCalibration=Hist. Binning, Confidence=Verbalized2026.0419
LLaVA-NeXT-34BCalibration=HAC-Gate, Confidence=Verbalized2026.0418
LLaVA-NeXT-34BCalibration=HAC-Platt, Confidence=Verbalized2026.0416.6
Qwen3-VL-2BCalibration=HAC-Gate, Confidence=Sampling2026.0414.4
Qwen3-VL-2BCalibration=HAC-Platt, Confidence=Verbalized2026.0414.3
Qwen3-VL-2BCalibration=HAC-Platt, Confidence=Sampling2026.0414
LLaVA-NeXT-7BCalibration=Platt Scaling, Confidence=Verbalized2026.0414
InternVL3-2BCalibration=HAC-Gate, Confidence=Verbalized2026.0413.9
InternVL3-38BCalibration=HAC-Gate, Confidence=Verbalized2026.0413.9
Qwen3-VL-2BCalibration=HAC-Gate, Confidence=Verbalized2026.0413.8
Qwen3-VL-32BCalibration=Hist. Binning, Confidence=Verbalized2026.0413.7
Qwen3-VL-32BCalibration=Platt Scaling, Confidence=Sampling2026.0413.5
Qwen3-VL-8BCalibration=Hist. Binning, Confidence=Sampling2026.0413.2
Qwen3-VL-32BCalibration=Hist. Binning, Confidence=Sampling2026.0413.2
Qwen3-VL-32BCalibration=HAC-Platt, Confidence=Sampling2026.0413.2
InternVL3-38BCalibration=Isotonic Regr., Confidence=Sampling2026.0413
Qwen3-VL-2BCalibration=Hist. Binning, Confidence=Verbalized2026.0412.8
Qwen3-VL-8BCalibration=Platt Scaling, Confidence=Sampling2026.0412.8
Qwen3-VL-8BCalibration=Platt Scaling, Confidence=Verbalized2026.0412.8
InternVL3-38BCalibration=HAC-Platt, Confidence=Verbalized2026.0412.7
LLaVA-NeXT-7BCalibration=Isotonic Regr., Confidence=Sampling2026.0412.7
Qwen3-VL-2BCalibration=Platt Scaling, Confidence=Sampling2026.0412.6
InternVL3-2BCalibration=HAC-Platt, Confidence=Verbalized2026.0412.6
Qwen3-VL-8BCalibration=Isotonic Regr., Confidence=Sampling2026.0412.5
InternVL3-8BCalibration=Isotonic Regr., Confidence=Sampling2026.0412.5
InternVL3-38BCalibration=Platt Scaling, Confidence=Sampling2026.0412.5
InternVL3-2BCalibration=Platt Scaling, Confidence=Verbalized2026.0412.4
InternVL3-8BCalibration=Hist. Binning, Confidence=Sampling2026.0412.4
InternVL3-38BCalibration=Hist. Binning, Confidence=Verbalized2026.0412.4
Qwen3-VL-8BCalibration=HAC-Gate, Confidence=Verbalized2026.0412.3
InternVL3-8BCalibration=Platt Scaling, Confidence=Verbalized2026.0412.3
Qwen3-VL-32BCalibration=HAC-Gate, Confidence=Sampling2026.0412.1
LLaVA-NeXT-7BCalibration=Isotonic Regr., Confidence=Verbalized2026.0412.1
Qwen3-VL-8BCalibration=Isotonic Regr., Confidence=Verbalized2026.0412
InternVL3-38BCalibration=HAC-Platt, Confidence=Sampling2026.0412
Qwen3-VL-2BCalibration=Hist. Binning, Confidence=Sampling2026.0411.9
InternVL3-2BCalibration=HAC-Platt, Confidence=Sampling2026.0411.9
LLaVA-NeXT-7BCalibration=HAC-Platt, Confidence=Verbalized2026.0411.8
Qwen3-VL-2BCalibration=Isotonic Regr., Confidence=Sampling2026.0411.7
Qwen3-VL-2BCalibration=Isotonic Regr., Confidence=Verbalized2026.0411.7
Qwen3-VL-32BCalibration=Isotonic Regr., Confidence=Sampling2026.0411.7
InternVL3-2BCalibration=Hist. Binning, Confidence=Sampling2026.0411.7
InternVL3-2BCalibration=Hist. Binning, Confidence=Verbalized2026.0411.7
InternVL3-2BCalibration=HAC-Gate, Confidence=Sampling2026.0411.7
InternVL3-8BCalibration=Platt Scaling, Confidence=Sampling2026.0411.7
InternVL3-38BCalibration=Platt Scaling, Confidence=Verbalized2026.0411.7
InternVL3-38BCalibration=Hist. Binning, Confidence=Sampling2026.0411.7
LLaVA-NeXT-7BCalibration=HAC-Gate, Confidence=Verbalized2026.0411.7
Qwen3-VL-2BCalibration=Platt Scaling, Confidence=Verbalized2026.0411.6
InternVL3-2BCalibration=Platt Scaling, Confidence=Sampling2026.0411.6
InternVL3-38BCalibration=Isotonic Regr., Confidence=Verbalized2026.0411.6
LLaVA-NeXT-7BCalibration=Hist. Binning, Confidence=Sampling2026.0411.6
LLaVA-NeXT-7BCalibration=HAC-Gate, Confidence=Sampling2026.0411.6
InternVL3-8BCalibration=Hist. Binning, Confidence=Verbalized2026.0411.5
LLaVA-NeXT-7BCalibration=HAC-Platt, Confidence=Sampling2026.0411.5
Qwen3-VL-8BCalibration=HAC-Gate, Confidence=Sampling2026.0411.4
Qwen3-VL-32BCalibration=Platt Scaling, Confidence=Verbalized2026.0411.4
InternVL3-2BCalibration=Isotonic Regr., Confidence=Sampling2026.0411.4
InternVL3-8BCalibration=Isotonic Regr., Confidence=Verbalized2026.0411.4
InternVL3-2BCalibration=Isotonic Regr., Confidence=Verbalized2026.0411.3
InternVL3-8BCalibration=HAC-Gate, Confidence=Sampling2026.0411.3
LLaVA-NeXT-7BCalibration=Hist. Binning, Confidence=Verbalized2026.0411.2
Qwen3-VL-8BCalibration=HAC-Platt, Confidence=Sampling2026.0411.1
Qwen3-VL-8BCalibration=HAC-Platt, Confidence=Verbalized2026.0411.1
LLaVA-NeXT-7BCalibration=Platt Scaling, Confidence=Sampling2026.0411.1
InternVL3-8BCalibration=HAC-Platt, Confidence=Sampling2026.0411
InternVL3-38BCalibration=HAC-Gate, Confidence=Sampling2026.0411
Qwen3-VL-32BCalibration=HAC-Gate, Confidence=Verbalized2026.0410.9
InternVL3-8BCalibration=HAC-Gate, Confidence=Verbalized2026.0410.5
Qwen3-VL-8BCalibration=Hist. Binning, Confidence=Verbalized2026.0410.4
Qwen3-VL-32BCalibration=HAC-Platt, Confidence=Verbalized2026.0410.4
Qwen3-VL-32BCalibration=Isotonic Regr., Confidence=Verbalized2026.0410
InternVL3-8BCalibration=HAC-Platt, Confidence=Verbalized2026.0410