English

$\mathbf{(N,K)}$-Puzzle: A Cost-Efficient Testbed for Benchmarking Reinforcement Learning Algorithms in Generative Language Model

Machine Learning 2024-03-13 v1 Artificial Intelligence Computation and Language

Abstract

Recent advances in reinforcement learning (RL) algorithms aim to enhance the performance of language models at scale. Yet, there is a noticeable absence of a cost-effective and standardized testbed tailored to evaluating and comparing these algorithms. To bridge this gap, we present a generalized version of the 24-Puzzle: the (N,K)(N,K)-Puzzle, which challenges language models to reach a target value KK with NN integers. We evaluate the effectiveness of established RL algorithms such as Proximal Policy Optimization (PPO), alongside novel approaches like Identity Policy Optimization (IPO) and Direct Policy Optimization (DPO).

Keywords

Cite

@article{arxiv.2403.07191,
  title  = {$\mathbf{(N,K)}$-Puzzle: A Cost-Efficient Testbed for Benchmarking Reinforcement Learning Algorithms in Generative Language Model},
  author = {Yufeng Zhang and Liyu Chen and Boyi Liu and Yingxiang Yang and Qiwen Cui and Yunzhe Tao and Hongxia Yang},
  journal= {arXiv preprint arXiv:2403.07191},
  year   = {2024}
}

Comments

8 pages

R2 v1 2026-06-28T15:16:31.688Z