@inproceedings{foosherian-etal-2023-enhancing,
title = "Enhancing Pipeline-Based Conversational Agents with Large Language Models",
author = "Foosherian, Mina and
Purwins, Hendrik and
Rathnayake, Purna and
Alam, Touhidul and
Teimao, Rui and
Thoben, Klaus-Dieter",
editor = "Hazarika, Devamanyu and
Tang, Xiangru Robert and
Jin, Di",
booktitle = "Proceedings of the 1st Workshop on Taming Large Language Models: Controllability in the era of Interactive Assistants!",
month = sep,
year = "2023",
address = "Prague, Czech Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.tllm-1.6",
pages = "56--67",
abstract = "The latest advancements in AI and deep learning have led to a breakthrough in large language model (LLM)-based agents such as GPT-4. However, many commercial conversational agent development tools are pipeline-based and have limitations in holding a human-like conversation. This paper investigates the capabilities of LLMs to enhance pipeline-based conversational agents during two phases: 1) in the design and development phase and 2) during operations. In 1) LLMs can aid in generating training data, extracting entities and synonyms, localization, and persona design. In 2) LLMs can assist in contextualization, intent classification to prevent conversational breakdown and handle out-of-scope questions, auto-correcting utterances, rephrasing responses, formulating disambiguation questions, summarization, and enabling closed question-answering capabilities. We conducted informal experiments with GPT-4 in the private banking domain to demonstrate the scenarios above with a practical example. Companies may be hesitant to replace their pipeline-based agents with LLMs entirely due to privacy concerns and the need for deep integration within their existing ecosystems. A hybrid approach in which LLMs{'} are integrated into the pipeline-based agents allows them to save time and costs of building and running agents by capitalizing on the capabilities of LLMs while retaining the integration and privacy safeguards of their existing systems.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="foosherian-etal-2023-enhancing">
<titleInfo>
<title>Enhancing Pipeline-Based Conversational Agents with Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mina</namePart>
<namePart type="family">Foosherian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hendrik</namePart>
<namePart type="family">Purwins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Purna</namePart>
<namePart type="family">Rathnayake</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Touhidul</namePart>
<namePart type="family">Alam</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Teimao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Klaus-Dieter</namePart>
<namePart type="family">Thoben</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-09</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on Taming Large Language Models: Controllability in the era of Interactive Assistants!</title>
</titleInfo>
<name type="personal">
<namePart type="given">Devamanyu</namePart>
<namePart type="family">Hazarika</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiangru</namePart>
<namePart type="given">Robert</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Prague, Czech Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The latest advancements in AI and deep learning have led to a breakthrough in large language model (LLM)-based agents such as GPT-4. However, many commercial conversational agent development tools are pipeline-based and have limitations in holding a human-like conversation. This paper investigates the capabilities of LLMs to enhance pipeline-based conversational agents during two phases: 1) in the design and development phase and 2) during operations. In 1) LLMs can aid in generating training data, extracting entities and synonyms, localization, and persona design. In 2) LLMs can assist in contextualization, intent classification to prevent conversational breakdown and handle out-of-scope questions, auto-correcting utterances, rephrasing responses, formulating disambiguation questions, summarization, and enabling closed question-answering capabilities. We conducted informal experiments with GPT-4 in the private banking domain to demonstrate the scenarios above with a practical example. Companies may be hesitant to replace their pipeline-based agents with LLMs entirely due to privacy concerns and the need for deep integration within their existing ecosystems. A hybrid approach in which LLMs’ are integrated into the pipeline-based agents allows them to save time and costs of building and running agents by capitalizing on the capabilities of LLMs while retaining the integration and privacy safeguards of their existing systems.</abstract>
<identifier type="citekey">foosherian-etal-2023-enhancing</identifier>
<location>
<url>https://aclanthology.org/2023.tllm-1.6</url>
</location>
<part>
<date>2023-09</date>
<extent unit="page">
<start>56</start>
<end>67</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Enhancing Pipeline-Based Conversational Agents with Large Language Models
%A Foosherian, Mina
%A Purwins, Hendrik
%A Rathnayake, Purna
%A Alam, Touhidul
%A Teimao, Rui
%A Thoben, Klaus-Dieter
%Y Hazarika, Devamanyu
%Y Tang, Xiangru Robert
%Y Jin, Di
%S Proceedings of the 1st Workshop on Taming Large Language Models: Controllability in the era of Interactive Assistants!
%D 2023
%8 September
%I Association for Computational Linguistics
%C Prague, Czech Republic
%F foosherian-etal-2023-enhancing
%X The latest advancements in AI and deep learning have led to a breakthrough in large language model (LLM)-based agents such as GPT-4. However, many commercial conversational agent development tools are pipeline-based and have limitations in holding a human-like conversation. This paper investigates the capabilities of LLMs to enhance pipeline-based conversational agents during two phases: 1) in the design and development phase and 2) during operations. In 1) LLMs can aid in generating training data, extracting entities and synonyms, localization, and persona design. In 2) LLMs can assist in contextualization, intent classification to prevent conversational breakdown and handle out-of-scope questions, auto-correcting utterances, rephrasing responses, formulating disambiguation questions, summarization, and enabling closed question-answering capabilities. We conducted informal experiments with GPT-4 in the private banking domain to demonstrate the scenarios above with a practical example. Companies may be hesitant to replace their pipeline-based agents with LLMs entirely due to privacy concerns and the need for deep integration within their existing ecosystems. A hybrid approach in which LLMs’ are integrated into the pipeline-based agents allows them to save time and costs of building and running agents by capitalizing on the capabilities of LLMs while retaining the integration and privacy safeguards of their existing systems.
%U https://aclanthology.org/2023.tllm-1.6
%P 56-67
Markdown (Informal)
[Enhancing Pipeline-Based Conversational Agents with Large Language Models](https://aclanthology.org/2023.tllm-1.6) (Foosherian et al., TLLM-WS 2023)
ACL
- Mina Foosherian, Hendrik Purwins, Purna Rathnayake, Touhidul Alam, Rui Teimao, and Klaus-Dieter Thoben. 2023. Enhancing Pipeline-Based Conversational Agents with Large Language Models. In Proceedings of the 1st Workshop on Taming Large Language Models: Controllability in the era of Interactive Assistants!, pages 56–67, Prague, Czech Republic. Association for Computational Linguistics.