Guha Balakrishnan
1 indexed paper
Recent (6 mo)
1With code
0Influential cites
0Benchmarked
0Publications per year
126
Top categories
AI×1NLP×1ML×1Stats ML×1
Frequent co-authors
Research Timeline
2026
Conformal Certification of Reasoning Trace Prefixes
The paper introduces CROP, a novel conformal procedure that provides rigorous statistical guarantees for certifying the longest safe prefix of a language model's reasoning trace, allowing for targeted error identification and repair.
Highlighted terms show continued research focus across papers