Nothing Special   »   [go: up one dir, main page]

Learning Task Decomposition to Assist Humans in Competitive Programming

Jiaxin Wen, Ruiqi Zhong, Pei Ke, Zhihong Shao, Hongning Wang, Minlie Huang


Abstract
When using language models (LMs) to solve complex problems, humans might struggle to understand the LM-generated solutions and repair the flawed ones. To assist humans in repairing them, we propose to automatically decompose complex solutions into multiple simpler pieces that correspond to specific subtasks. We introduce a novel objective for learning task decomposition, termed assistive value (AssistV), which measures the feasibility and speed for humans to repair the decomposed solution. We collect a dataset of human repair experiences on different decomposed solutions. Utilizing the collected data as in-context examples, we then learn to critique, refine, and rank decomposed solutions to improve AssistV. We validate our method under competitive programming problems: under 177 hours of human study, our method enables non-experts to solve 33.3% more problems, speeds them up by 3.3x, and empowers them to match unassisted experts.
Anthology ID:
2024.acl-long.629
Volume:
Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
August
Year:
2024
Address:
Bangkok, Thailand
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
11700–11723
Language:
URL:
https://aclanthology.org/2024.acl-long.629
DOI:
10.18653/v1/2024.acl-long.629
Bibkey:
Cite (ACL):
Jiaxin Wen, Ruiqi Zhong, Pei Ke, Zhihong Shao, Hongning Wang, and Minlie Huang. 2024. Learning Task Decomposition to Assist Humans in Competitive Programming. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11700–11723, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Learning Task Decomposition to Assist Humans in Competitive Programming (Wen et al., ACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.acl-long.629.pdf