paper-with-me

홈 › Papers

ToRL: Scaling Tool-Integrated RL

2025-03-30 · Xuefeng Li, Haoyang Zou, PengFei Liu

We introduce ToRL (Tool-Integrated Reinforcement Learning), a framework for training large language models (LLMs) to autonomously use computational tools via reinforcement learning. Unlike supervised fine-tuning, ToRL allows models to explore and discover optimal strategies for tool use. Experiments with Qwen2.5-Math models show significant improvements: ToRL-7B reaches 43.3\% accuracy on AIME~24, surpassing reinforcement learning without tool integration by 14\% and the best existing Tool-Integrated Reasoning (TIR) model by 17\%. Further analysis reveals emergent behaviors such as strategic tool invocation, self-regulation of ineffective code, and dynamic adaptation between computational and analytical reasoning, all arising purely through reward-driven learning.

📄 PDF Abstract BibTeX arXiv:2503.23383

Code (1)

gair-nlp/torl pytorch

Tasks

Mathreinforcement-learningReinforcement Learning

Similar Papers 제목 키워드 기반

PowerNet: Transferable Dynamic IR Drop Estimation via Maximum Convolutional Neural Network

2020-11-26 · Zhiyao Xie, Haoxing Ren, Brucek Khailany, Ye Sheng 외

IR drop is a fundamental constraint required by almost all chip designs. However, its evaluation usually takes a long time that hinders mitigation techniques for fixing its violations. In this work, we develop a fast dyn…

AutoRL Hyperparameter Landscapes

2023-04-05 · Aditya Mohan, Carolin Benjamins, Konrad Wienecke, Alexander Dockhorn 외

Although Reinforcement Learning (RL) has shown to be capable of producing impressive results, its use is limited by the impact of its hyperparameters on performance. This often makes it difficult to achieve good results …

AutoMLHyperparameter OptimizationOpen-Ended Question AnsweringReinforcement Learning (RL)

A Scalable Reinforcement Learning-based System Using On-Chain Data for Cryptocurrency Portfolio Management

2023-07-04 · Zhenhan Huang, Fumihide Tanaka

On-chain data (metrics) of blockchain networks, akin to company fundamentals, provide crucial and comprehensive insights into the networks. Despite their informative nature, on-chain data have not been utilized in reinfo…

ManagementReinforcement Learning (RL)

Corpora and Processing Tools for Non-standard Contemporary and Diachronic Balkan Slavic

2019-09-01 · RANLP 2019 9 · Teodora Vukovic, Nora Muheim, Olivier Winist{\"o}rfer, Ivan {\v{S}}imko 외

The paper describes three corpora of different varieties of BS that are currently being developed with the goal of providing data for the analysis of the diatopic and diachronic variation in non-standard Balkan Slavic. T…

LemmatizationPOS

FactorLLM: Factorizing Knowledge via Mixture of Experts for Large Language Models

2024-08-15 · Zhongyu Zhao, Menghang Dong, Rongyu Zhang, Wenzhao Zheng 외

Recent research has demonstrated that Feed-Forward Networks (FFNs) in Large Language Models (LLMs) play a pivotal role in storing diverse linguistic and factual knowledge. Conventional methods frequently face challenges …

Mixture-of-Experts