@inproceedings{pu-etal-2023-just,
title = "Just Adjust One Prompt: Enhancing In-Context Dialogue Scoring via Constructing the Optimal Subgraph of Demonstrations and Prompts",
author = "Pu, Jiashu and
Cheng, Ling and
Fan, Lu and
Lv, Tangjie and
Zhang, Rongsheng",
editor = "Bouamor, Houda and
Pino, Juan and
Bali, Kalika",
booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.emnlp-main.590/",
doi = "10.18653/v1/2023.emnlp-main.590",
pages = "9472--9496",
abstract = "The use of modern Large Language Models (LLMs) as chatbots still has some problems such as hallucinations and lack of empathy. Identifying these issues can help improve chatbot performance. The community has been continually iterating on reference-free dialogue evaluation methods based on large language models (LLMs) that can be readily applied. However, many of these LLM-based metrics require selecting specific datasets and developing specialized training tasks for different evaluation dimensions (e.g., coherence, informative). The developing step can be time-consuming and may need to be repeated for new evaluation dimensions. To enable efficient and flexible adaptation to diverse needs of dialogue evaluation, we propose a dimension-agnostic scoring method that leverages the in-context learning (ICL) capability of LLMs to learn from human scoring to the fullest extent. Our method has three key features. To begin with, rather than manual prompt crafting, we propose automatically generating prompts, allowing the LLM to observe human labels and summarize the most suitable prompt. Additionally, since the LLM has a token limit and ICL is sensitive to demonstration variations, we train a selector to finely customize demonstrations and prompts for each dialogue input. Finally, during inference, we propose to request the LLM multiple times with a subgraph of demonstrations and prompts that are diverse and suitable to maximize ICL from various human scoring. We validate the efficacy of our method on five datasets, even with a small amount of annotated data, our method outperforms all strong baselines. Code is available at https://github.com/iamlxb3/EMNLP2023-ADOROR."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pu-etal-2023-just">
<titleInfo>
<title>Just Adjust One Prompt: Enhancing In-Context Dialogue Scoring via Constructing the Optimal Subgraph of Demonstrations and Prompts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jiashu</namePart>
<namePart type="family">Pu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ling</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Fan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tangjie</namePart>
<namePart type="family">Lv</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rongsheng</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Bouamor</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="family">Pino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalika</namePart>
<namePart type="family">Bali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The use of modern Large Language Models (LLMs) as chatbots still has some problems such as hallucinations and lack of empathy. Identifying these issues can help improve chatbot performance. The community has been continually iterating on reference-free dialogue evaluation methods based on large language models (LLMs) that can be readily applied. However, many of these LLM-based metrics require selecting specific datasets and developing specialized training tasks for different evaluation dimensions (e.g., coherence, informative). The developing step can be time-consuming and may need to be repeated for new evaluation dimensions. To enable efficient and flexible adaptation to diverse needs of dialogue evaluation, we propose a dimension-agnostic scoring method that leverages the in-context learning (ICL) capability of LLMs to learn from human scoring to the fullest extent. Our method has three key features. To begin with, rather than manual prompt crafting, we propose automatically generating prompts, allowing the LLM to observe human labels and summarize the most suitable prompt. Additionally, since the LLM has a token limit and ICL is sensitive to demonstration variations, we train a selector to finely customize demonstrations and prompts for each dialogue input. Finally, during inference, we propose to request the LLM multiple times with a subgraph of demonstrations and prompts that are diverse and suitable to maximize ICL from various human scoring. We validate the efficacy of our method on five datasets, even with a small amount of annotated data, our method outperforms all strong baselines. Code is available at https://github.com/iamlxb3/EMNLP2023-ADOROR.</abstract>
<identifier type="citekey">pu-etal-2023-just</identifier>
<identifier type="doi">10.18653/v1/2023.emnlp-main.590</identifier>
<location>
<url>https://aclanthology.org/2023.emnlp-main.590/</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>9472</start>
<end>9496</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Just Adjust One Prompt: Enhancing In-Context Dialogue Scoring via Constructing the Optimal Subgraph of Demonstrations and Prompts
%A Pu, Jiashu
%A Cheng, Ling
%A Fan, Lu
%A Lv, Tangjie
%A Zhang, Rongsheng
%Y Bouamor, Houda
%Y Pino, Juan
%Y Bali, Kalika
%S Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F pu-etal-2023-just
%X The use of modern Large Language Models (LLMs) as chatbots still has some problems such as hallucinations and lack of empathy. Identifying these issues can help improve chatbot performance. The community has been continually iterating on reference-free dialogue evaluation methods based on large language models (LLMs) that can be readily applied. However, many of these LLM-based metrics require selecting specific datasets and developing specialized training tasks for different evaluation dimensions (e.g., coherence, informative). The developing step can be time-consuming and may need to be repeated for new evaluation dimensions. To enable efficient and flexible adaptation to diverse needs of dialogue evaluation, we propose a dimension-agnostic scoring method that leverages the in-context learning (ICL) capability of LLMs to learn from human scoring to the fullest extent. Our method has three key features. To begin with, rather than manual prompt crafting, we propose automatically generating prompts, allowing the LLM to observe human labels and summarize the most suitable prompt. Additionally, since the LLM has a token limit and ICL is sensitive to demonstration variations, we train a selector to finely customize demonstrations and prompts for each dialogue input. Finally, during inference, we propose to request the LLM multiple times with a subgraph of demonstrations and prompts that are diverse and suitable to maximize ICL from various human scoring. We validate the efficacy of our method on five datasets, even with a small amount of annotated data, our method outperforms all strong baselines. Code is available at https://github.com/iamlxb3/EMNLP2023-ADOROR.
%R 10.18653/v1/2023.emnlp-main.590
%U https://aclanthology.org/2023.emnlp-main.590/
%U https://doi.org/10.18653/v1/2023.emnlp-main.590
%P 9472-9496
Markdown (Informal)
[Just Adjust One Prompt: Enhancing In-Context Dialogue Scoring via Constructing the Optimal Subgraph of Demonstrations and Prompts](https://aclanthology.org/2023.emnlp-main.590/) (Pu et al., EMNLP 2023)
ACL