Let's Verify Step by Step

• URL: http://arxiv.org/abs/2305.20050v1
• Date: Wed, 31 May 2023 17:24:00 GMT
• Title: Let's Verify Step by Step
• Authors: Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, Karl Cobbe
• Abstract summary: We show that process supervision significantly outperforms outcome supervision for training models to solve problems. Our model solves 78% of problems from a representative subset of the MATH test set. We also release PRM800K, the complete dataset of 800,000 step-level human feedback labels used to train our best reward model.
• Score: 73.58107073356732
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In recent years, large language models have greatly improved in their ability to perform complex multi-step reasoning. However, even state-of-the-art models still regularly produce logical mistakes. To train more reliable models, we can turn either to outcome supervision, which provides feedback for a final result, or process supervision, which provides feedback for each intermediate reasoning step. Given the importance of training reliable models, and given the high cost of human feedback, it is important to carefully compare the both methods. Recent work has already begun this comparison, but many questions still remain. We conduct our own investigation, finding that process supervision significantly outperforms outcome supervision for training models to solve problems from the challenging MATH dataset. Our process-supervised model solves 78% of problems from a representative subset of the MATH test set. Additionally, we show that active learning significantly improves the efficacy of process supervision. To support related research, we also release PRM800K, the complete dataset of 800,000 step-level human feedback labels used to train our best reward model.

Related papers

• Guiding Through Complexity: What Makes Good Supervision for Hard Reasoning Tasks? [74.88417042125985]
  We investigate various data-driven strategies that offer supervision data at different quality levels upon tasks of varying complexity. We find that even when the outcome error rate for hard task supervision is high, training on such data can outperform perfectly correct supervision on easier subtasks. Our results also reveal that supplementing hard task supervision with the corresponding subtask supervision can yield notable performance improvements.
  arXiv  Detail & Related papers  (2024-10-27T17:55:27Z)
• Self-Taught Evaluators [77.92610887220594]
  We present an approach that aims to im-proves without human annotations, using synthetic training data only. Our Self-Taught Evaluator can improve a strong LLM from 75.4 to 88.3 on RewardBench.
  arXiv  Detail & Related papers  (2024-08-05T17:57:02Z)
• Improve Mathematical Reasoning in Language Models by Automated Process Supervision [22.72856086318912]
  We propose a novel Monte Carlo Tree Search (MCTS) algorithm named textitOmegaPRM for the efficient collection of high-quality process supervision data. We are able to collect over 1.5 million process supervision annotations to train a Process Reward Model (PRM) We have enhanced the instruction tuned Gemini Pro model's math reasoning performance, achieving a 69.4% success rate on the MATH benchmark.
  arXiv  Detail & Related papers  (2024-06-05T19:25:40Z)
• Multi-step Problem Solving Through a Verifier: An Empirical Analysis on Model-induced Process Supervision [40.984680166762345]
  We introduce Model-induced Process Supervision (MiPS), a novel method for automating data curation. MiPS annotates an intermediate step by sampling completions of this solution through the reasoning model, and obtaining an accuracy defined as the proportion of correct completions. Our approach significantly improves the performance of PaLM 2 on math and coding tasks.
  arXiv  Detail & Related papers  (2024-02-05T00:57:51Z)
• Teaching Language Models to Self-Improve through Interactive Demonstrations [83.9421355808174]
  Self-improving ability of large language models has been shown to be absent and difficult to learn for smaller models. We introduce TriPosT, a training algorithm that endows smaller models with such self-improvement ability. We show that our approach can improve a LLaMA-7b's performance on math and reasoning tasks by up to 7.13%.
  arXiv  Detail & Related papers  (2023-10-20T14:11:04Z)
• AdaMerging: Adaptive Model Merging for Multi-Task Learning [68.75885518081357]
  This paper introduces an innovative technique called Adaptive Model Merging (AdaMerging) It aims to autonomously learn the coefficients for model merging, either in a task-wise or layer-wise manner, without relying on the original training data. Compared to the current state-of-the-art task arithmetic merging scheme, AdaMerging showcases a remarkable 11% improvement in performance.
  arXiv  Detail & Related papers  (2023-10-04T04:26:33Z)
• Learning to Modulate pre-trained Models in RL [22.812215561012874]
  Fine-tuning a pre-trained model often suffers from catastrophic forgetting. Our study shows that with most fine-tuning approaches, the performance on pre-training tasks deteriorates significantly. We propose a novel method, Learning-to-Modulate (L2M), that avoids the degradation of learned skills by modulating the information flow of the frozen pre-trained model.
  arXiv  Detail & Related papers  (2023-06-26T17:53:05Z)
• SUPERB-SG: Enhanced Speech processing Universal PERformance Benchmark for Semantic and Generative Capabilities [76.97949110580703]
  We introduce SUPERB-SG, a new benchmark to evaluate pre-trained models across various speech tasks. We use a lightweight methodology to test the robustness of representations learned by pre-trained models under shifts in data domain. We also show that the task diversity of SUPERB-SG coupled with limited task supervision is an effective recipe for evaluating the generalizability of model representation.
  arXiv  Detail & Related papers  (2022-03-14T04:26:40Z)
• Training Verifiers to Solve Math Word Problems [12.307284507186342]
  We introduce GSM8K, a dataset of 8.5K high quality linguistically diverse grade school math word problems. We find that even the largest transformer models fail to achieve high test performance. To increase performance, we propose training verifiers to judge the correctness of model completions.
  arXiv  Detail & Related papers  (2021-10-27T04:49:45Z)
• Sample Efficient Reinforcement Learning via Model-Ensemble Exploration and Exploitation [3.728946517493471]
  MEEE is a model-ensemble method that consists of optimistic exploration and weighted exploitation. Our approach outperforms other model-free and model-based state-of-the-art methods, especially in sample complexity.
  arXiv  Detail & Related papers  (2021-07-05T07:18:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.