ADeptS-Bench - Disambiguation (Mobile): leaderboard
Metric: Disambiguation F1 (%; harmonic mean of the precision and recall of the clarification questions against human ground-truth items; 381 mobile tasks; ambiguous GUI decision points reached through a trajectory of screenshots; generated clarification questions matched to ground-truth items by a Gemini 3.1 Pro judge; with severity scoring prompt). Source: arxiv.org. Saturation forecast: Around May 2028. 7 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3.1 Pro (Preview) | 58.9 |
| 2 | Claude Opus 4.7 | 53.7 |
| 3 | Qwen 3 VL 235B A22B | 52.8 |
| 4 | GPT-5.4 | 45 |
Interactive version: theaggregate.ai/benchmark?slug=adepts-bench-disambiguation-mobile · How It Works · Data refreshed daily, snapshot 2026-09-29.