Medical Visual Question Answering on VQA-Med Closed
0.013ECEInternVL3-38B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| InternVL3-38BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.013 | — | |
| InternVL3-38BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.013 | — | |
| Qwen3-VL-32BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.021 | — | |
| Qwen3-VL-32BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.021 | — | |
| Qwen3-VL-32BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.022 | — | |
| Qwen3-VL-32BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.022 | — | |
| Qwen3-VL-32BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.022 | — | |
| InternVL3-8BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.024 | — | |
| InternVL3-8BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.025 | — | |
| InternVL3-8BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.026 | — | |
| InternVL3-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.027 | — | |
| InternVL3-8BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.028 | — | |
| Qwen3-VL-8BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.032 | — | |
| Qwen3-VL-32BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.036 | — | |
| Qwen3-VL-2BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.037 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.039 | — | |
| InternVL3-38BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.041 | — | |
| Qwen3-VL-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.042 | — | |
| Qwen3-VL-2BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.042 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.047 | — | |
| LLaVA-NeXT-7BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.048 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.048 | — | |
| LLaVA-NeXT-34BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.049 | — | |
| LLaVA-NeXT-34BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.05 | — | |
| Qwen3-VL-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.051 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.051 | — | |
| Qwen3-VL-2BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.061 | — | |
| Qwen3-VL-2BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.062 | — | |
| Qwen3-VL-2BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.062 | — | |
| Qwen3-VL-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.063 | — | |
| Qwen3-VL-2BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.064 | — | |
| Qwen3-VL-8BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.064 | — | |
| Qwen3-VL-2BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.065 | — | |
| Qwen3-VL-8BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.065 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.065 | — | |
| InternVL3-38BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.069 | — | |
| LLaVA-NeXT-7BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.069 | — | |
| Qwen3-VL-32BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.077 | — | |
| InternVL3-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.077 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.078 | — | |
| Qwen3-VL-8BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.079 | — | |
| Qwen3-VL-8BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.081 | — | |
| Qwen3-VL-2BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.083 | — | |
| Qwen3-VL-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.085 | — | |
| Qwen3-VL-8BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.087 | — | |
| InternVL3-38BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.087 | — | |
| Qwen3-VL-32BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.088 | — | |
| InternVL3-8BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.089 | — | |
| InternVL3-2BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.09 | — | |
| Qwen3-VL-8BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.091 | — | |
| InternVL3-2BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.096 | — | |
| InternVL3-2BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.096 | — | |
| InternVL3-8BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.101 | — | |
| InternVL3-8BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.102 | — | |
| LLaVA-NeXT-7BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.104 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.105 | — | |
| LLaVA-NeXT-34BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.106 | — | |
| InternVL3-2BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.107 | — | |
| InternVL3-38BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.109 | — | |
| Qwen3-VL-32BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.113 | — | |
| InternVL3-8BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.113 | — | |
| InternVL3-38BCalibration Strategy=Platt Scaling, Confidence Mode=Verbalized2026.04 | 0.113 | — | |
| InternVL3-2BCalibration Strategy=HAC-Gate, Confidence Mode=Verbalized2026.04 | 0.114 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Isotonic Regr., Confidence Mode=Verbalized2026.04 | 0.114 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.116 | — | |
| InternVL3-8BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.128 | — | |
| Qwen3-VL-8BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.129 | — | |
| InternVL3-38BCalibration Strategy=HAC-Platt, Confidence Mode=Verbalized2026.04 | 0.13 | — | |
| InternVL3-2BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.136 | — | |
| Qwen3-VL-32BCalibration Strategy=Hist. Binning, Confidence Mode=Verbalized2026.04 | 0.138 | — | |
| InternVL3-8BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.146 | — | |
| InternVL3-2BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.151 | — | |
| InternVL3-2BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.154 | — | |
| Qwen3-VL-8BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.159 | — | |
| InternVL3-38BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.162 | — | |
| InternVL3-8BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.163 | — | |
| LLaVA-NeXT-7BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.169 | — | |
| InternVL3-2BCalibration Strategy=Platt Scaling, Confidence Mode=Sampling2026.04 | 0.178 | — | |
| InternVL3-2BCalibration Strategy=HAC-Platt, Confidence Mode=Sampling2026.04 | 0.178 | — | |
| InternVL3-38BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.178 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Isotonic Regr., Confidence Mode=Sampling2026.04 | 0.181 | — | |
| LLaVA-NeXT-34BCalibration Strategy=HAC-Gate, Confidence Mode=Sampling2026.04 | 0.185 | — | |
| Qwen3-VL-32BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.2 | — | |
| InternVL3-2BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.21 | — | |
| InternVL3-38BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.215 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.218 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Hist. Binning, Confidence Mode=Sampling2026.04 | 0.226 | — | |
| Qwen3-VL-8BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.242 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.258 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.265 | — | |
| Qwen3-VL-2BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.286 | — | |
| LLaVA-NeXT-34BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.291 | — | |
| Qwen3-VL-32BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.296 | — | |
| Qwen3-VL-2BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.297 | — | |
| InternVL3-38BCalibration Strategy=Uncalibrated, Confidence Mode=Sampling2026.04 | 0.338 | — | |
| LLaVA-NeXT-7BCalibration Strategy=Uncalibrated, Confidence Mode=Verbalized2026.04 | 0.453 | — | |
| 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 |