Evaluating the Role of Verifiers in Test-Time Scaling for Legal Reasoning Tasks
This work addresses the underexplored application of test-time scaling in legal domains, providing insights for improving LLM performance in argumentative tasks, though it is incremental as it builds on existing TTS techniques.
The study tackled the problem of evaluating verifier-based test-time scaling methods for legal reasoning tasks, finding that domain specialization and supervision type significantly affect verifier utility across five benchmarks.
Test-time scaling (TTS) techniques can improve the performance of large language models (LLMs) at the expense of additional computation and latency. While TTS has proven effective in formal domains such as mathematics and programming, its value in argumentative domains such as law remains underexplored. We present an empirical study of verifier-based TTS methods for legal multiple-choice QA (MCQA) across five benchmarks. Using a family of 7 reward models, we evaluate both outcome-level (Best-of-$N$) and process-level (tree search) verification under realistic low-$N$ budgets. Our analysis systematically investigates how verifier utility is affected by key properties such as domain specialization, model size, and supervision type (process-supervised PRMs vs. outcome-only ORMs), even when applied across different roles.