Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs

Large Language Models (LLMs), such as ChatGPT, greatly empower dialogue systems with strong language understanding and generation capabilities. However, most of the previous works prompt the LLMs to directly generate a response based on the dialogue context, overlooking the underlying linguistic cue...

Full description

Saved in:
Bibliographic Details
Main Authors: WANG, Hongru, WANG, Rui, MI, Fei, DENG, Yang, WANG, Zezhong, LIANG, Bin, XU, Ruifeng, WONG, Kam-Fai
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2023
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/9122
https://ink.library.smu.edu.sg/context/sis_research/article/10125/viewcontent/2023.findings_emnlp.806.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-10125
record_format dspace
spelling sg-smu-ink.sis_research-101252024-08-01T14:34:25Z Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs WANG, Hongru WANG, Rui MI, Fei DENG, Yang WANG, Zezhong LIANG, Bin XU, Ruifeng WONG, Kam-Fai Large Language Models (LLMs), such as ChatGPT, greatly empower dialogue systems with strong language understanding and generation capabilities. However, most of the previous works prompt the LLMs to directly generate a response based on the dialogue context, overlooking the underlying linguistic cues about the user status exhibited in the context. Such in-depth dialogue scenarios are challenging for existing LLMs to figure out the user’s hidden needs and respond satisfactorily through a single-step inference. To this end, we propose a novel linguistic cue-based chain-of-thoughts (Cue-CoT), which enhances the LLMs inference with an intermediate reasoning step to find cues exhibited in the dialogue, aiming to provide a more personalized and engaging response. To evaluate the approach, we build a benchmark with in-depth dialogue questions, consisting of 6 datasets in both Chinese and English, targeting 3 major linguistic cues during the conversation: personality, emotion, and psychology. We conducted experiments on the proposed benchmark with 5 LLMs under both zero-shot and one-shot settings. Empirical results demonstrate our proposed Cue-CoT method outperforms standard prompting methods in terms of both helpfulness and acceptability on all datasets. 2023-12-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/9122 info:doi/10.18653/v1/2023.findings-emnlp.806 https://ink.library.smu.edu.sg/context/sis_research/article/10125/viewcontent/2023.findings_emnlp.806.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Databases and Information Systems
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Databases and Information Systems
spellingShingle Databases and Information Systems
WANG, Hongru
WANG, Rui
MI, Fei
DENG, Yang
WANG, Zezhong
LIANG, Bin
XU, Ruifeng
WONG, Kam-Fai
Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs
description Large Language Models (LLMs), such as ChatGPT, greatly empower dialogue systems with strong language understanding and generation capabilities. However, most of the previous works prompt the LLMs to directly generate a response based on the dialogue context, overlooking the underlying linguistic cues about the user status exhibited in the context. Such in-depth dialogue scenarios are challenging for existing LLMs to figure out the user’s hidden needs and respond satisfactorily through a single-step inference. To this end, we propose a novel linguistic cue-based chain-of-thoughts (Cue-CoT), which enhances the LLMs inference with an intermediate reasoning step to find cues exhibited in the dialogue, aiming to provide a more personalized and engaging response. To evaluate the approach, we build a benchmark with in-depth dialogue questions, consisting of 6 datasets in both Chinese and English, targeting 3 major linguistic cues during the conversation: personality, emotion, and psychology. We conducted experiments on the proposed benchmark with 5 LLMs under both zero-shot and one-shot settings. Empirical results demonstrate our proposed Cue-CoT method outperforms standard prompting methods in terms of both helpfulness and acceptability on all datasets.
format text
author WANG, Hongru
WANG, Rui
MI, Fei
DENG, Yang
WANG, Zezhong
LIANG, Bin
XU, Ruifeng
WONG, Kam-Fai
author_facet WANG, Hongru
WANG, Rui
MI, Fei
DENG, Yang
WANG, Zezhong
LIANG, Bin
XU, Ruifeng
WONG, Kam-Fai
author_sort WANG, Hongru
title Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs
title_short Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs
title_full Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs
title_fullStr Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs
title_full_unstemmed Cue-CoT: Chain-of-thought prompting for responding to in-depth dialogue questions with LLMs
title_sort cue-cot: chain-of-thought prompting for responding to in-depth dialogue questions with llms
publisher Institutional Knowledge at Singapore Management University
publishDate 2023
url https://ink.library.smu.edu.sg/sis_research/9122
https://ink.library.smu.edu.sg/context/sis_research/article/10125/viewcontent/2023.findings_emnlp.806.pdf
_version_ 1814047748298637312