Year: 2025
Type: conference-paper
Abstract: Reinforcement learning (RL) has shown promise in enhancing large language model (LLM) reasoning, yet progress towards broader capabilities is limited by the availability of high-quality, multi-domain datasets.This work introduces GURU, a 92K RL-for-reasoning dataset designed to address this gap, covering six reasoning domains: Math, Code, Science, Logic, Simulation, and Tabular, each with corresponding verifiers.We build GURU via a careful data-curation pipeline, including sourcing, deduplication, reward design, and domain-specific and difficulty-based filtering.With GURU, we present a systematic investigation of cross-domain RL generalization, and reveal several key aspects affecting crossdomain transferability.We further train two models GURU-7B and GURU-32B purely with RL on our curated data and observe largely improved performance over leading open RL reasoning model baselines, with gains of 7.3% and 7.8% respectively on an extensive 17-task, six-domain evaluation suite.We are releasing our dataset, code, and evaluation suite to the community, aiming to support further research and development of more general RL-enhanced reasoning models.
DOI: 10.52202/085713-4479
Authors: Jorge (Zhoujun) Cheng, Shibo Hao, Tianyang Liu, Fan Zhou, Yutao Xie, Feng Yao, Yuexin Bian, Nilabjo Dey, Yonghao Zhuang, Yuheng Zha, Yi Gu, Kun Zhou, Yuqi Wang, Yuan Li, Richard Fan, Jianshu She, Chengqian Gao, Abulhair Saparov, Taylor W. Killian, Haonan Li, Mikhail Yurochkin, Eric Xing, Zhengzhong Liu, Zhiting Hu
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7197020508
