V-STaR: Training Verifiers for Self-Taught Reasoners
Hosseini, Arian, Yuan, Xingdi, Malkin, Nikolay, Courville, Aaron, Sordoni, Alessandro, Agarwal, Rishabh
–arXiv.org Artificial Intelligence
Common self-improvement approaches for large language models (LLMs), such as STaR (Zelikman et al., 2022), iteratively fine-tune LLMs on self-generated solutions to improve their problem-solving ability. However, these approaches discard the large amounts of incorrect solutions generated during this process, potentially neglecting valuable information in such solutions. To address this shortcoming, we propose V-STaR that utilizes both the correct and incorrect solutions generated during the self-improvement process to train a verifier using DPO that judges correctness of model-generated solutions. This verifier is used at inference time to select one solution among many candidate solutions. Running V-STaR for multiple iterations results in progressively better reasoners and verifiers, delivering a 4% to 17% test accuracy improvement over existing self-improvement and verification approaches on common code generation and math reasoning benchmarks with LLaMA2 models.
arXiv.org Artificial Intelligence
Feb-9-2024
- Country:
- North America > Canada (0.28)
- Genre:
- Research Report > New Finding (0.46)
- Industry:
- Education (0.68)
- Technology: