The authors focus on question-answering for LRMs, where alignment via reinforcement learning tends to produce systematic overconfidence and suppress output variability.
Jailbreak prompts improve uncertainty calibration in black-box reasoning models
The authors show that prompt-level relaxation via jailbreaks broadens output distributions and yields significantly better-calibrated confidence than existing black-box UQ baselines.
Research Lab
Lucas Biechy · Cédric Eichler · Adrien Boiret · Nicolas Anciaux
Petscraft · Inria · Université Paris-Saclay · INSA CVL · Université d'Orléans
Research Digest··1 min read
Biechy et al.
Why this paper
From Inria and 5 others
In one line
Jailbreak prompts, repurposed for uncertainty estimation, significantly improve confidence calibration in black-box large reasoning models.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§