Enhancing Reinforcement Learning with Label-Sensitive Reward for Natural Language Understanding

Kuo Liao, Shuang Li, Meng Zhao, Liqun Liu, Mengge Xue, Zhenyu Hu, Honglin Han, Chengguo Yin


Abstract
Recent strides in large language models (LLMs) have yielded remarkable performance, leveraging reinforcement learning from human feedback (RLHF) to significantly enhance generation and alignment capabilities. However, RLHF encounters numerous challenges, including the objective mismatch issue, leading to suboptimal performance in Natural Language Understanding (NLU) tasks.To address this limitation, we propose a novel Reinforcement Learning framework enhanced with Label-sensitive Reward (RLLR) to amplify the performance of LLMs in NLU tasks. By incorporating label-sensitive pairs into reinforcement learning, our method aims to adeptly capture nuanced label-sensitive semantic features during RL, thereby enhancing natural language understanding.Experiments conducted on five diverse foundation models across eight tasks showcase promising results. In comparison to Supervised Fine-tuning models (SFT), RLLR demonstrates an average performance improvement of 1.54%. Compared with RLHF models, the improvement averages at 0.69%. These results reveal the effectiveness of our method for LLMs in NLU tasks.
Anthology ID:
2024.acl-long.231
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:
4206–4220
Language:
URL:
https://aclanthology.org/2024.acl-long.231
DOI:
10.18653/v1/2024.acl-long.231
Bibkey:
Cite (ACL):
Kuo Liao, Shuang Li, Meng Zhao, Liqun Liu, Mengge Xue, Zhenyu Hu, Honglin Han, and Chengguo Yin. 2024. Enhancing Reinforcement Learning with Label-Sensitive Reward for Natural Language Understanding. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4206–4220, Bangkok, Thailand. Association for Computational Linguistics.
Cite (Informal):
Enhancing Reinforcement Learning with Label-Sensitive Reward for Natural Language Understanding (Liao et al., ACL 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.acl-long.231.pdf