Instruct Once, Chat Consistently in Multiple Rounds: An Efficient Tuning Framework for Dialogue
Jian Wang, Chak Tou Leong, Jiashuo Wang, Dongding Lin, Wenjie Li, Xiaoyong Wei
Abstract
Tuning language models for dialogue generation has been a prevalent paradigm for building capable dialogue agents. Yet, traditional tuning narrowly views dialogue generation as resembling other language generation tasks, ignoring the role disparities between two speakers and the multi-round interactive process that dialogues ought to be. Such a manner often leads to unsatisfactory chat consistency for the built agent. In this work, we emphasize the interactive, communicative nature of dialogue and argue that it is more feasible to model the speaker roles of agent and user separately, enabling the agent to adhere to its role consistently. With this in mind, we propose an efficient Multi-round Interactive Dialogue Tuning (Midi-Tuning) framework. It models the agent and user individually with two adapters built upon large language models. The adapters make use of respective utterances round by round in alternating order and they are tuned via a round-level memory caching mechanism. Extensive experiments demonstrate that, our framework performs superior to traditional fine-tuning and harbors the tremendous potential for improving dialogue consistency.- Anthology ID:
- 2024.acl-long.219
- 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:
- 3993–4010
- Language:
- URL:
- https://aclanthology.org/2024.acl-long.219
- DOI:
- Cite (ACL):
- Jian Wang, Chak Tou Leong, Jiashuo Wang, Dongding Lin, Wenjie Li, and Xiaoyong Wei. 2024. Instruct Once, Chat Consistently in Multiple Rounds: An Efficient Tuning Framework for Dialogue. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 3993–4010, Bangkok, Thailand. Association for Computational Linguistics.
- Cite (Informal):
- Instruct Once, Chat Consistently in Multiple Rounds: An Efficient Tuning Framework for Dialogue (Wang et al., ACL 2024)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2024.acl-long.219.pdf