Step-by-step verifiers — also known as process reward models (PMR) — are a
key ingredient for test-time scaling. PRMs require step-level supervision,
making them expensive to train. This work aims to build data-efficient PRMs as
verbalized step-wise reward models that verify every step in the solution by
generating a verification chain-of-thought (CoT). We propose ThinkPRM, a long
CoT verifier fine-tuned on orders of magnitude fewer process labels than those
required by discriminative PRMs. Our approach capitalizes on the inherent
reasoning abilities of long CoT models, and outperforms LLM-as-a-Judge and
discriminative verifiers — using only 1% of the process labels in PRM800K —
across several challenging benchmarks. Spécifiquement, ThinkPRM beats the
baselines on ProcessBench, MATH-500, and AIME ’24 under best-of-N selection and
reward-guided search. In an out-of-domain evaluation on a subset of
GPQA-Diamond and LiveCodeBench, our PRM surpasses discriminative verifiers
trained on the full PRM800K by 8% et 4.5%, respectively. Lastly, under the
same token budget, ThinkPRM scales up verification compute more effectively
compared to LLM-as-a-Judge, outperforming it by 7.2% on a subset of
ProcessBench. Our work highlights the value of generative, long CoT PRMs that
can scale test-time compute for verification while requiring minimal
supervision for training. Our code, données, and models will be released at
https://github.com/mukhal/thinkprm.

Cet article explore les excursions dans le temps et leurs implications.

Télécharger PDF: