Coffee-Gym: An Environment for Evaluating and Improving Natural Language Feedback on Erroneous Code

Hyungjoo Chae, Taeyoon Kwon, Seung Jun Moon, Yongho Song, Dongjin Kang, kai tzu-iunn ong, Beong Woo Kwak, Seonghyeon Bae, Seung-won Hwang, Jinyoung Yeo

Coffee-Gym: An Environment for Evaluating and Improving Natural Language Feedback on Erroneous Code: 10 upvotes on Hugging Face Daily Papers, #10 of 13 papers on 2024-10-01. Day-by-day upvote history.

This paper presents Coffee-Gym, a comprehensive RL environment for training models that provide feedback on code editing. Coffee-Gym includes two major components: (1) Coffee, a dataset containing humans' code edit traces for coding questions and machine-written feedback for editing erroneous code; (2) CoffeeEval, a reward function that faithfully reflects the helpfulness of feedback by assessing the performance of the revised code in unit tests. With them, Coffee-Gym addresses the unavailability of high-quality datasets for training feedback models with RL, and provides more accurate rewards than the SOTA reward model (i.e., GPT-4). By applying Coffee-Gym, we elicit feedback models that outperform baselines in enhancing open-source code LLMs' code editing, making them comparable with closed-source LLMs. We make the dataset and the model checkpoint publicly available.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.