CoT-Self-Instruct: Building high-quality synthetic prompts for reasoning and non-reasoning tasks

• URL: http://arxiv.org/abs/2507.23751v1
• Date: Thu, 31 Jul 2025 17:38:50 GMT
• Title: CoT-Self-Instruct: Building high-quality synthetic prompts for reasoning and non-reasoning tasks
• Authors: Ping Yu, Jack Lanchantin, Tianlu Wang, Weizhe Yuan, Olga Golovneva, Ilia Kulikov, Sainbayar Sukhbaatar, Jason Weston, Jing Xu,
• Abstract summary: We propose CoT-Self-Instruct, a synthetic data generation method that instructs LLMs to first reason and plan via Chain-of-Thought (CoT)<n>In verifiable reasoning, our synthetic data significantly outperforms existing training datasets, such as s1k and OpenMathReasoning.<n>For non-verifiable instruction-following tasks, our method surpasses the performance of human or standard self-instruct prompts on both AlpacaEval 2.0 and Arena-Hard.
• Score: 57.482238100217195
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We propose CoT-Self-Instruct, a synthetic data generation method that instructs LLMs to first reason and plan via Chain-of-Thought (CoT) based on the given seed tasks, and then to generate a new synthetic prompt of similar quality and complexity for use in LLM training, followed by filtering for high-quality data with automatic metrics. In verifiable reasoning, our synthetic data significantly outperforms existing training datasets, such as s1k and OpenMathReasoning, across MATH500, AMC23, AIME24 and GPQA-Diamond. For non-verifiable instruction-following tasks, our method surpasses the performance of human or standard self-instruct prompts on both AlpacaEval 2.0 and Arena-Hard.

Related papers

• UnitCoder: Scalable Iterative Code Synthesis with Unit Test Guidance [65.01483640267885]
  Large Language Models (LLMs) have demonstrated remarkable capabilities in various tasks, yet code generation remains a major challenge.<n>We introduce UnitCoder, a systematic pipeline leveraging model-generated unit tests to guide and validate the code generation process.<n>Our work presents a scalable approach that leverages model-generated unit tests to guide the synthesis of high-quality code data from pre-training corpora.
  arXiv  Detail & Related papers  (2025-02-17T05:37:02Z)
• Synthesizing Post-Training Data for LLMs through Multi-Agent Simulation [51.20656279478878]
  MATRIX is a multi-agent simulator that automatically generates diverse text-based scenarios.<n>We introduce MATRIX-Gen for controllable and highly realistic data synthesis.<n>On AlpacaEval 2 and Arena-Hard benchmarks, Llama-3-8B-Base, post-trained on datasets synthesized by MATRIX-Gen with just 20K instruction-response pairs, outperforms Meta's Llama-3-8B-Instruct model.
  arXiv  Detail & Related papers  (2024-10-18T08:01:39Z)
• Optimizing Instruction Synthesis: Effective Exploration of Evolutionary Space with Tree Search [25.108044778194536]
  We introduce IDEA-MCTS (Instruction Data Enhancement using Monte Carlo Tree Search), a scalable framework for efficiently synthesizing instructions. With tree search and evaluation models, it can efficiently guide each instruction to evolve into a high-quality form, aiding in instruction fine-tuning. Experimental results show that IDEA-MCTS significantly enhances the seed instruction data, raising the average evaluation scores of quality, diversity, and complexity from 2.19 to 3.81.
  arXiv  Detail & Related papers  (2024-10-14T11:28:30Z)
• Genetic Instruct: Scaling up Synthetic Generation of Coding Instructions for Large Language Models [59.60208063956459]
  Large Language Models (LLMs) require high quality instruction data for effective alignment.<n>We present Genetic-Instruct, a scalable algorithm for synthesizing large-scale, high quality coding instructions.
  arXiv  Detail & Related papers  (2024-07-29T20:42:59Z)
• CodecLM: Aligning Language Models with Tailored Synthetic Data [51.59223474427153]
  We introduce CodecLM, a framework for adaptively generating high-quality synthetic data for instruction-following abilities. We first encode seed instructions into metadata, which are concise keywords generated on-the-fly to capture the target instruction distribution. We also introduce Self-Rubrics and Contrastive Filtering during decoding to tailor data-efficient samples.
  arXiv  Detail & Related papers  (2024-04-08T21:15:36Z)
• Improving Text Embeddings with Large Language Models [59.930513259982725]
  We introduce a novel and simple method for obtaining high-quality text embeddings using only synthetic data and less than 1k training steps. We leverage proprietary LLMs to generate diverse synthetic data for hundreds of thousands of text embedding tasks across 93 languages. Experiments demonstrate that our method achieves strong performance on highly competitive text embedding benchmarks without using any labeled data.
  arXiv  Detail & Related papers  (2023-12-31T02:13:18Z)
• Reinforcement Learning and Data-Generation for Syntax-Guided Synthesis [0.0]
  We present a reinforcement-learning algorithm for SyGuS which uses Monte-Carlo Tree Search (MCTS) to search the space of candidate solutions. Our algorithm learns policy and value functions which, combined with the upper confidence bound for trees, allow it to balance exploration and exploitation.
  arXiv  Detail & Related papers  (2023-07-13T11:30:50Z)
• Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction [28.51694365908817]
  This work shows that useful data can be synthetically generated even for tasks that cannot be solved directly by large language models. We synthetically generate a dataset of 1.8M data points, establish its superior quality compared to existing datasets in a human evaluation.
  arXiv  Detail & Related papers  (2023-03-07T18:48:55Z)

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.