Cohere Labs' CALIBER trains models to estimate and revise confidence around reasoning
Cohere_Labs · x · 2026-10-21
Cohere Labs published a research blog introducing CALIBER, based on the observation that confidence before and after reasoning answer two different questions: 'How likely am I to solve this?' versus 'How likely is this answer to be correct?'
The work trains models to estimate both forms of confidence and to revise their confidence after reasoning, aiming to make models' self-assessments better calibrated to actual performance. Details are in the Cohere Labs blog.
More from Research
- Anima Anandkumar on neural operators: AI weather models beat supercomputers — AnimaAnandkumar · 2026-10-23
- Scaling GLM-5.3-Flash training to 1M token context, 64x over Megatron baseline — AndrewLBeam · 2026-10-23
- Experimental boolean-logic LM: microsecond forward pass on one CPU thread, but painfully slow training — cephaloform · 2026-10-23
- AdaDex turns seconds of human hand motion into generalizable dexterous robot skills — mangahomanga · 2026-10-23
- ReAPO paper: conditioning retries on failures beats GRPO in RLVR reasoning training — jasondeanlee · 2026-10-23
- Quantum-JEPA: first JEPA world model deployable on real quantum circuits (QPU) — DrMorganLevine · 2026-10-23