Symbolic Planning and Code Generation for Grounded Dialogue
Justin Chiu, Wenting Zhao, Derek Chen, Saujas Vaduguru, Alexander Rush, Daniel Fried
Abstract
Large language models (LLMs) excel at processing and generating both text and code. However, LLMs have had limited applicability in grounded task-oriented dialogue as they are difficult to steer toward task objectives and fail to handle novel grounding. We present a modular and interpretable grounded dialogue system that addresses these shortcomings by composing LLMs with a symbolic planner and grounded code execution. Our system consists of a reader and planner: the reader leverages an LLM to convert partner utterances into executable code, calling functions that perform grounding. The translated code’s output is stored to track dialogue state, while a symbolic planner determines the next appropriate response. We evaluate our system’s performance on the demanding OneCommon dialogue task, involving collaborative reference resolution on abstract images of scattered dots. Our system substantially outperforms the previous state-of-the-art, including improving task success in human evaluations from 56% to 69% in the most challenging setting.- Anthology ID:
- 2023.pandl-1.5
- Volume:
- Proceedings of the 2nd Workshop on Pattern-based Approaches to NLP in the Age of Deep Learning
- Month:
- December
- Year:
- 2023
- Address:
- Singapore
- Editors:
- Mihai Surdeanu, Ellen Riloff, Laura Chiticariu, Dayne Frietag, Gus Hahn-Powell, Clayton T. Morrison, Enrique Noriega-Atala, Rebecca Sharp, Marco Valenzuela-Escarcega
- Venues:
- PANDL | WS
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 43–53
- Language:
- URL:
- https://aclanthology.org/2023.pandl-1.5
- DOI:
- 10.18653/v1/2023.pandl-1.5
- Cite (ACL):
- Justin Chiu, Wenting Zhao, Derek Chen, Saujas Vaduguru, Alexander Rush, and Daniel Fried. 2023. Symbolic Planning and Code Generation for Grounded Dialogue. In Proceedings of the 2nd Workshop on Pattern-based Approaches to NLP in the Age of Deep Learning, pages 43–53, Singapore. Association for Computational Linguistics.
- Cite (Informal):
- Symbolic Planning and Code Generation for Grounded Dialogue (Chiu et al., PANDL-WS 2023)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2023.pandl-1.5.pdf