RoboAbstention - False Premise: leaderboard
The False Premise category of RoboAbstention (2026, The Yes-Man Syndrome): 957 instructions built on a false premise about the scene, given with an image to a vision-language robot planner that should abstain or ask. The score is the abstention rate in percent; higher is better.
Metric: Abstention rate (%) on the 957 False Premise instructions, image plus instruction to an embodied VLM planner, every instruction warranting abstention, act or abstain judged by GPT-5.4 mini (97.5 percent agreement with human labels), API default sampling; higher is better. Source: arxiv.org. Saturation forecast: Not forecast. 8 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Setting | Score | Overall rank |
|---|---|---|---|---|
| 1 | Gemini 2.5 Flash | 28.5 | #463 | |
| 2 | Llama 4 Maverick | 15.5 | #1110 | |
| 3 | Qwen 3.5 27B | 7.6 | #375 | |
| 4 | Claude Sonnet 4.6 | 7.2 | #173 | |
| 5 | Gemini Robotics-ER 1.6 | 7 | #170 | |
| 6 | GPT-5.4 | 3.6 | #135 | |
| 7 | GPT-5.4 Mini | High | 3.4 | #406 |
| 8 | GPT-5.4 Nano | 1.5 | #651 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=roboabstention-false-premise · How It Works · Data refreshed daily, snapshot 2026-11-21.