Evaluating Large Language Models for Document-grounded Response Generation in Information-Seeking Dialogues

Norbert Braunschweiler, Rama Doddipatla, Simon Keizer, Svetlana Stoyanchev


Abstract
In this paper, we investigate the use of large language models (LLMs) like ChatGPT for document-grounded response generation in the context of information-seeking dialogues. For evaluation, we use the MultiDoc2Dial corpus of task-oriented dialogues in four social service domains previously used in the DialDoc 2022 Shared Task. Information-seeking dialogue turns are grounded in multiple documents providing relevant information. We generate dialogue completion responses by prompting a ChatGPT model, using two methods: Chat-Completion and LlamaIndex. ChatCompletion uses knowledge from ChatGPT model pre-training while LlamaIndex also extracts relevant information from documents. Observing that document-grounded response generation via LLMs cannot be adequately assessed by automatic evaluation metrics as they are significantly more verbose, we perform a human evaluation where annotators rate the output of the shared task winning system, the two ChatGPT variants outputs, and human responses. While both ChatGPT variants are more likely to include information not present in the relevant segments, possibly including a presence of hallucinations, they are rated higher than both the shared task winning system and human responses.
Anthology ID:
2023.tllm-1.5
Volume:
Proceedings of the 1st Workshop on Taming Large Language Models: Controllability in the era of Interactive Assistants!
Month:
September
Year:
2023
Address:
Prague, Czech Republic
Editors:
Devamanyu Hazarika, Xiangru Robert Tang, Di Jin
Venues:
TLLM | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
46–55
Language:
URL:
https://aclanthology.org/2023.tllm-1.5
DOI:
Bibkey:
Cite (ACL):
Norbert Braunschweiler, Rama Doddipatla, Simon Keizer, and Svetlana Stoyanchev. 2023. Evaluating Large Language Models for Document-grounded Response Generation in Information-Seeking Dialogues. In Proceedings of the 1st Workshop on Taming Large Language Models: Controllability in the era of Interactive Assistants!, pages 46–55, Prague, Czech Republic. Association for Computational Linguistics.
Cite (Informal):
Evaluating Large Language Models for Document-grounded Response Generation in Information-Seeking Dialogues (Braunschweiler et al., TLLM-WS 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.tllm-1.5.pdf