List your company

Visual Question Answering on VQA-RAD Closed

70.2AUROCInternVL3-2B
43.88850.71957.5564.381Apr 2, 2026

Evaluation results

MethodLinks
InternVL3-2BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0470.2
InternVL3-2BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0469.7
InternVL3-2BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0468.7
InternVL3-2BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0468.7
InternVL3-2BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0468.1
InternVL3-2BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0468
InternVL3-2BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0467.5
InternVL3-2BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0467.5
InternVL3-2BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0467
InternVL3-38BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0466.6
InternVL3-38BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0466.6
InternVL3-38BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0466.6
InternVL3-38BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0466.6
InternVL3-2BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0466.3
InternVL3-2BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0466.2
InternVL3-38BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0464.5
InternVL3-2BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0464.3
LLaVA-NeXT-34BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0462.5
LLaVA-NeXT-34BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0462.5
LLaVA-NeXT-34BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0462.4
LLaVA-NeXT-34BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0462.4
Qwen3-VL-2BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0462.1
Qwen3-VL-2BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0462.1
LLaVA-NeXT-7BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0461.8
LLaVA-NeXT-7BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0461.8
LLaVA-NeXT-7BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0461.8
LLaVA-NeXT-7BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0461.8
InternVL3-38BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0461.7
InternVL3-8BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0461.6
InternVL3-8BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0461.6
Qwen3-VL-2BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0461.2
InternVL3-8BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0461
LLaVA-NeXT-34BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0460.7
Qwen3-VL-2BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0460.1
InternVL3-8BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0460
InternVL3-38BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0460
LLaVA-NeXT-34BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0460
InternVL3-38BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0459.8
InternVL3-38BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0459.8
InternVL3-38BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0459.8
InternVL3-8BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0459.5
Qwen3-VL-8BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0459.2
Qwen3-VL-8BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0459.1
Qwen3-VL-32BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0458.8
Qwen3-VL-32BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0458.8
LLaVA-NeXT-7BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0458.8
Qwen3-VL-8BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0458
Qwen3-VL-32BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0457.6
LLaVA-NeXT-7BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0457.6
Qwen3-VL-8BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0457.5
Qwen3-VL-8BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0457.5
InternVL3-8BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0457
Qwen3-VL-2BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0456.6
Qwen3-VL-8BCalibration Method=HAC-Platt, Confidence Type=Sampling2026.0456.6
LLaVA-NeXT-7BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0456.5
Qwen3-VL-8BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0456.2
InternVL3-38BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0456.2
Qwen3-VL-2BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0456.1
LLaVA-NeXT-7BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0456.1
Qwen3-VL-2BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0456
Qwen3-VL-2BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0456
Qwen3-VL-32BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0455.9
Qwen3-VL-32BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0455.9
Qwen3-VL-32BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0455.4
LLaVA-NeXT-34BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0455.3
Qwen3-VL-32BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0455.1
Qwen3-VL-32BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0455.1
LLaVA-NeXT-7BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0455.1
LLaVA-NeXT-34BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0455.1
LLaVA-NeXT-7BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0454.9
LLaVA-NeXT-7BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0454.9
LLaVA-NeXT-34BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0454.9
LLaVA-NeXT-34BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0454.9
LLaVA-NeXT-34BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0454.8
Qwen3-VL-8BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0454.6
Qwen3-VL-32BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0454.1
Qwen3-VL-8BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0453.9
InternVL3-38BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0453.9
LLaVA-NeXT-34BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0453.8
Qwen3-VL-8BCalibration Method=Uncalibrated, Confidence Type=Sampling2026.0453.7
Qwen3-VL-8BCalibration Method=Platt Scaling, Confidence Type=Sampling2026.0453.7
Qwen3-VL-8BCalibration Method=HAC-Gate, Confidence Type=Sampling2026.0453.4
Qwen3-VL-32BCalibration Method=Isotonic Regr., Confidence Type=Sampling2026.0452.2
Qwen3-VL-2BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0451.8
Qwen3-VL-2BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0451.8
Qwen3-VL-2BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0451.5
Qwen3-VL-32BCalibration Method=Hist. Binning, Confidence Type=Sampling2026.0451.5
InternVL3-8BCalibration Method=HAC-Gate, Confidence Type=Verbalized2026.0451.2
Qwen3-VL-32BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0451.1
Qwen3-VL-2BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0450.4
InternVL3-8BCalibration Method=Uncalibrated, Confidence Type=Verbalized2026.0450.3
InternVL3-8BCalibration Method=Platt Scaling, Confidence Type=Verbalized2026.0450.3
InternVL3-8BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0449.9
LLaVA-NeXT-7BCalibration Method=HAC-Platt, Confidence Type=Verbalized2026.0448.2
InternVL3-8BCalibration Method=Isotonic Regr., Confidence Type=Verbalized2026.0447.6
InternVL3-8BCalibration Method=Hist. Binning, Confidence Type=Verbalized2026.0444.9