Expert clinical evaluation on 100 Cycles Expert Review
4.09AcceptabilityGPT-5.2 (VDSS)
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GPT-5.2 (VDSS)Framework=VDSS (multi agent), Backbone=GPT-5.22026.05 | 4.09 | 4.46 | 4.15 | 4.11 | |
| NoPref (Ablation)Framework=VDSS (multi agent), Backbone=GPT-5.2, Ablation=No Preference2026.05 | 3.81 | 4.32 | 3.95 | 3.89 | |
| Gemini 3 Pro (VDSS)Framework=VDSS (multi agent), Backbone=Gemini 3 Pro2026.05 | 3.8 | 4.3 | 3.91 | 3.82 | |
| NoImg (Ablation)Framework=VDSS (multi agent), Backbone=GPT-5.2, Ablation=No Image2026.05 | 3.69 | 4.26 | 3.75 | 3.62 | |
| Qwen3-VL-235B (VDSS)Framework=VDSS (multi agent), Backbone=Qwen3-VL-235B2026.05 | 3.51 | 4.2 | 3.79 | 3.59 | |
| NoImgNoPref (Ablation)Framework=VDSS (multi agent), Backbone=GPT-5.2, Ablation=No Image & No Preference2026.05 | 3.5 | 4.15 | 3.61 | 3.46 | |
| Qwen3(VL)-8B (VDSS)Framework=VDSS (multi agent), Backbone=Qwen3-VL-8B2026.05 | 3.3 | 4.12 | 3.56 | 3.22 | |
| Gemini 3 ProFramework=Single model end to end, Backbone=Gemini 3 Pro2026.05 | 2.8 | 2.31 | 2.24 | 2.45 | |
| GPT-5.2Framework=Single model end to end, Backbone=GPT-5.22026.05 | 2.75 | 2.43 | 2.71 | 2.63 | |
| Qwen3-VL-235BFramework=Single model end to end, Backbone=Qwen3-VL-235B2026.05 | 2.53 | 2.17 | 2.19 | 2.3 | |
| Qwen3-VL-8BFramework=Single model end to end, Backbone=Qwen3-VL-8B2026.05 | 2.45 | 2.11 | 2.26 | 2.27 |