RATIONALYST: Pre-training Process-Supervision for Improving Reasoning

• URL: http://arxiv.org/abs/2410.01044v1
• Date: Tue, 1 Oct 2024 20:05:51 GMT
• Title: RATIONALYST: Pre-training Process-Supervision for Improving Reasoning
• Authors: Dongwei Jiang, Guoxuan Wang, Yining Lu, Andrew Wang, Jingyu Zhang, Chuyu Liu, Benjamin Van Durme, Daniel Khashabi,
• Abstract summary: We introduce RATIONALYST, a model for process-supervision of reasoning based on pre-training. We extract 79k rationales from web-scale unlabelled dataset (the Pile) and a combination of reasoning datasets with minimal human intervention. Fine-tuned from LLaMa-3-8B, RATIONALYST improves the accuracy of reasoning by an average of 3.9% on 7 representative reasoning benchmarks.
• Score: 41.9992614617405
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The reasoning steps generated by LLMs might be incomplete, as they mimic logical leaps common in everyday communication found in their pre-training data: underlying rationales are frequently left implicit (unstated). To address this challenge, we introduce RATIONALYST, a model for process-supervision of reasoning based on pre-training on a vast collection of rationale annotations extracted from unlabeled data. We extract 79k rationales from web-scale unlabelled dataset (the Pile) and a combination of reasoning datasets with minimal human intervention. This web-scale pre-training for reasoning allows RATIONALYST to consistently generalize across diverse reasoning tasks, including mathematical, commonsense, scientific, and logical reasoning. Fine-tuned from LLaMa-3-8B, RATIONALYST improves the accuracy of reasoning by an average of 3.9% on 7 representative reasoning benchmarks. It also demonstrates superior performance compared to significantly larger verifiers like GPT-4 and similarly sized models fine-tuned on matching training sets.

Related papers

• P-FOLIO: Evaluating and Improving Logical Reasoning with Abundant Human-Written Reasoning Chains [97.25943550933829]
  We present P-FOLIO, a human-annotated dataset consisting of diverse and complex reasoning chains. We use P-FOLIO to evaluate and improve large-language-model (LLM) reasoning capabilities.
  arXiv  Detail & Related papers  (2024-10-11T19:22:57Z)
• ReGenesis: LLMs can Grow into Reasoning Generalists via Self-Improvement [70.09541267910974]
  Post-training Large Language Models (LLMs) with explicit reasoning trajectories can enhance their reasoning abilities. Existing self-synthesizing methods suffer from poor generalization to out-of-domain (OOD) reasoning tasks. We propose Reasoning Generalist via Self-Improvement (ReGenesis), a method to self-synthesize reasoning paths as post-training data.
  arXiv  Detail & Related papers  (2024-10-03T00:09:15Z)
• Evaluating Human Alignment and Model Faithfulness of LLM Rationale [66.75309523854476]
  We study how well large language models (LLMs) explain their generations through rationales. We show that prompting-based methods are less "faithful" than attribution-based explanations.
  arXiv  Detail & Related papers  (2024-06-28T20:06:30Z)
• General Purpose Verification for Chain of Thought Prompting [16.381123651223763]
  We explore ways to improve reasoning capabilities of Large Language Models (LLMs) We propose three general principles that a model should adhere to while reasoning. We apply these constraints to the reasoning steps generated by the LLM to improve the accuracy of the final generation.
  arXiv  Detail & Related papers  (2024-04-30T21:15:17Z)
• Learning Planning-based Reasoning by Trajectories Collection and Process Reward Synthesizing [61.98556945939045]
  We propose a framework to learn planning-based reasoning through Direct Preference Optimization (DPO) on collected trajectories. Our results on challenging logical reasoning benchmarks demonstrate the effectiveness of our learning framework.
  arXiv  Detail & Related papers  (2024-02-01T15:18:33Z)
• LogicAsker: Evaluating and Improving the Logical Reasoning Ability of Large Language Models [63.14196038655506]
  We introduce LogicAsker, a novel approach for evaluating and enhancing the logical reasoning capabilities of large language models (LLMs) Our methodology reveals significant gaps in LLMs' learning of logical rules, with identified reasoning failures ranging from 29% to 90% across different models. We leverage these findings to construct targeted demonstration examples and fine-tune data, notably enhancing logical reasoning in models like GPT-4o by up to 5%.
  arXiv  Detail & Related papers  (2024-01-01T13:53:53Z)
• REFINER: Reasoning Feedback on Intermediate Representations [47.36251998678097]
  We introduce REFINER, a framework for finetuning language models to generate intermediate inferences. REFINER works by interacting with a critic model that provides automated feedback on the reasoning. Empirical evaluations show significant improvements over baseline LMs of comparable scale.
  arXiv  Detail & Related papers  (2023-04-04T15:57:28Z)

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.