StanceBench - Assertiveness: leaderboard
StanceBench tests whether audio-capable models can read interpersonal stance from conversational speech: for each stance dimension, speakers were role-prompted toward a positive or a negative pole, and the judge must tell them apart from the audio. This board covers assertive and self-assured versus hesitant delivery. The score is the area under the ROC curve of the signed stance score the judge gives, against the role-derived poles, so 0.5 is chance.
Metric: AUROC (0-1 scale) of the judge's conversation-level signed stance score in separating the 91 role-prompted positive-pole and negative-pole single-speaker instances of the Assertiveness dimension (S3), 25 percent sample of the Seamless Interaction Improvised subset; binary pole choice asked in both pole orders, malformed outputs dropped. Source: arxiv.org. Saturation forecast: Around February 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 2.5 Flash (Thinking) | 0.87 | #453 (Gemini 2.5 Flash) |
| 2 | Qwen2.5-Omni-7B | 0.8 | #1789 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #5 Claude Opus 5.5.
Interactive version: theaggregate.ai/benchmark?slug=stancebench-assertiveness · How It Works · Data refreshed daily, snapshot 2026-11-13.