Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use

• URL: http://arxiv.org/abs/2504.04736v2
• Date: Mon, 28 Apr 2025 01:20:59 GMT
• Title: Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use
• Authors: Anna Goldie, Azalia Mirhoseini, Hao Zhou, Irene Cai, Christopher D. Manning,
• Abstract summary: Reinforcement learning has been shown to improve the performance of large language models.<n>We propose a synthetic data generation and RL methodology targeting multi-step optimization scenarios.<n>Step-Wise Reinforcement Learning (SWiRL) iteratively generates multi-step reasoning and tool use data, and then learns from that data.
• Score: 38.51141014728375
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning has been shown to improve the performance of large language models. However, traditional approaches like RLHF or RLAIF treat the problem as single-step. As focus shifts toward more complex reasoning and agentic tasks, language models must take multiple steps of text generation, reasoning and environment interaction before generating a solution. We propose a synthetic data generation and RL methodology targeting multi-step optimization scenarios. This approach, called Step-Wise Reinforcement Learning (SWiRL), iteratively generates multi-step reasoning and tool use data, and then learns from that data. It employs a simple step-wise decomposition that breaks each multi-step trajectory into multiple sub-trajectories corresponding to each action by the original model. It then applies synthetic data filtering and RL optimization on these sub-trajectories. We evaluated SWiRL on a number of multi-step tool use, question answering, and mathematical reasoning tasks. Our experiments show that SWiRL outperforms baseline approaches by 21.5%, 12.3%, 14.8%, 11.1%, and 15.3% in relative accuracy on GSM8K, HotPotQA, CofCA, MuSiQue, and BeerQA, respectively. Excitingly, the approach exhibits generalization across tasks: for example, training only on HotPotQA (text question-answering) improves zero-shot performance on GSM8K (a math dataset) by a relative 16.9%.

Related papers

• Nemotron-CrossThink: Scaling Self-Learning beyond Math Reasoning [66.43194385702297]
  Large Language Models (LLMs) have shown strong reasoning capabilities, particularly when enhanced through Reinforcement Learning (RL) We propose NEMOTRON-CROSSTHINK, a framework that systematically incorporates multi-domain corpora, including both synthetic and real-world question-answer pairs, into RL training to improve generalization across diverse reasoning tasks.
  arXiv  Detail & Related papers  (2025-04-15T21:37:13Z)
• LIMR: Less is More for RL Scaling [25.477841726836836]
  We introduce Learning Impact Measurement (LIM), an automated method to evaluate and prioritize training samples.<n>Our method achieves comparable or even superior performance using only 1,389 samples versus the full 8,523 samples dataset.<n>For reproducible research and future innovation, we are open-sourcing LIMR, including implementation of LIM, training and evaluation code, curated datasets, and trained models.
  arXiv  Detail & Related papers  (2025-02-17T15:13:29Z)
• Multi-Agent Sampling: Scaling Inference Compute for Data Synthesis with Tree Search-Based Agentic Collaboration [81.45763823762682]
  This work aims to bridge the gap by investigating the problem of data synthesis through multi-agent sampling.<n>We introduce Tree Search-based Orchestrated Agents(TOA), where the workflow evolves iteratively during the sequential sampling process.<n>Our experiments on alignment, machine translation, and mathematical reasoning demonstrate that multi-agent sampling significantly outperforms single-agent sampling as inference compute scales.
  arXiv  Detail & Related papers  (2024-12-22T15:16:44Z)
• Q-SFT: Q-Learning for Language Models via Supervised Fine-Tuning [62.984693936073974]
  Value-based reinforcement learning can learn effective policies for a wide range of multi-turn problems.<n>Current value-based RL methods have proven particularly challenging to scale to the setting of large language models.<n>We propose a novel offline RL algorithm that addresses these drawbacks, casting Q-learning as a modified supervised fine-tuning problem.
  arXiv  Detail & Related papers  (2024-11-07T21:36:52Z)
• ARLBench: Flexible and Efficient Benchmarking for Hyperparameter Optimization in Reinforcement Learning [42.33815055388433]
  ARLBench is a benchmark for hyperparameter optimization (HPO) in reinforcement learning (RL) It allows comparisons of diverse HPO approaches while being highly efficient in evaluation. ARLBench is an efficient, flexible, and future-oriented foundation for research on AutoRL.
  arXiv  Detail & Related papers  (2024-09-27T15:22:28Z)
• Unlock the Correlation between Supervised Fine-Tuning and Reinforcement Learning in Training Code Large Language Models [12.656574142412484]
  We make an attempt to understand the correlation between supervised fine-tuning and reinforcement learning.<n>We find that both atomic and synthetic functions are indispensable for SFT's generalization.
  arXiv  Detail & Related papers  (2024-06-14T03:39:01Z)
• An Empirical Study of Multimodal Model Merging [148.48412442848795]
  Model merging is a technique that fuses multiple models trained on different tasks to generate a multi-task solution. We conduct our study for a novel goal where we can merge vision, language, and cross-modal transformers of a modality-specific architecture. We propose two metrics that assess the distance between weights to be merged and can serve as an indicator of the merging outcomes.
  arXiv  Detail & Related papers  (2023-04-28T15:43:21Z)
• An Experimental Design Perspective on Model-Based Reinforcement Learning [73.37942845983417]
  In practical applications of RL, it is expensive to observe state transitions from the environment. We propose an acquisition function that quantifies how much information a state-action pair would provide about the optimal solution to a Markov decision process.
  arXiv  Detail & Related papers  (2021-12-09T23:13:57Z)

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.