@inproceedings{gupta-etal-2024-evaluating,
title = "Evaluating Robustness of Open Dialogue Summarization Models in the Presence of Naturally Occurring Variations",
author = "Gupta, Ankita and
Gunasekara, Chulaka and
Wan, Hui and
Ganhotra, Jatin and
Joshi, Sachindra and
Danilevsky, Marina",
editor = "Nouri, Elnaz and
Rastogi, Abhinav and
Spithourakis, Georgios and
Liu, Bing and
Chen, Yun-Nung and
Li, Yu and
Albalak, Alon and
Wakaki, Hiromi and
Papangelis, Alexandros",
booktitle = "Proceedings of the 6th Workshop on NLP for Conversational AI (NLP4ConvAI 2024)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.nlp4convai-1.4",
pages = "56--72",
abstract = "Dialogue summarization involves summarizing long conversations while preserving the most salient information. Real-life dialogues often involve naturally occurring variations (e.g., repetitions, hesitations). In this study, we systematically investigate the impact of such variations on state-of-the-art open dialogue summarization models whose details are publicly known (e.g., architectures, weights, and training corpora). To simulate real-life variations, we introduce two types of perturbations: utterance-level perturbations that modify individual utterances with errors and language variations, and dialogue-level perturbations that add non-informative exchanges (e.g., repetitions, greetings). We perform our analysis along three dimensions of robustness: consistency, saliency, and faithfulness, which aim to capture different aspects of performance of a summarization model. We find that both fine-tuned and instruction-tuned models are affected by input variations, with the latter being more susceptible, particularly to dialogue-level perturbations. We also validate our findings via human evaluation. Finally, we investigate whether the robustness of fine-tuned models can be improved by training them with a fraction of perturbed data. We find that this approach does not yield consistent performance gains, warranting further research. Overall, our work highlights robustness challenges in current open encoder-decoder summarization models and provides insights for future research.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gupta-etal-2024-evaluating">
<titleInfo>
<title>Evaluating Robustness of Open Dialogue Summarization Models in the Presence of Naturally Occurring Variations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ankita</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chulaka</namePart>
<namePart type="family">Gunasekara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hui</namePart>
<namePart type="family">Wan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jatin</namePart>
<namePart type="family">Ganhotra</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sachindra</namePart>
<namePart type="family">Joshi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marina</namePart>
<namePart type="family">Danilevsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 6th Workshop on NLP for Conversational AI (NLP4ConvAI 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Elnaz</namePart>
<namePart type="family">Nouri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Abhinav</namePart>
<namePart type="family">Rastogi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Georgios</namePart>
<namePart type="family">Spithourakis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bing</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alon</namePart>
<namePart type="family">Albalak</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hiromi</namePart>
<namePart type="family">Wakaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandros</namePart>
<namePart type="family">Papangelis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Dialogue summarization involves summarizing long conversations while preserving the most salient information. Real-life dialogues often involve naturally occurring variations (e.g., repetitions, hesitations). In this study, we systematically investigate the impact of such variations on state-of-the-art open dialogue summarization models whose details are publicly known (e.g., architectures, weights, and training corpora). To simulate real-life variations, we introduce two types of perturbations: utterance-level perturbations that modify individual utterances with errors and language variations, and dialogue-level perturbations that add non-informative exchanges (e.g., repetitions, greetings). We perform our analysis along three dimensions of robustness: consistency, saliency, and faithfulness, which aim to capture different aspects of performance of a summarization model. We find that both fine-tuned and instruction-tuned models are affected by input variations, with the latter being more susceptible, particularly to dialogue-level perturbations. We also validate our findings via human evaluation. Finally, we investigate whether the robustness of fine-tuned models can be improved by training them with a fraction of perturbed data. We find that this approach does not yield consistent performance gains, warranting further research. Overall, our work highlights robustness challenges in current open encoder-decoder summarization models and provides insights for future research.</abstract>
<identifier type="citekey">gupta-etal-2024-evaluating</identifier>
<location>
<url>https://aclanthology.org/2024.nlp4convai-1.4</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>56</start>
<end>72</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Evaluating Robustness of Open Dialogue Summarization Models in the Presence of Naturally Occurring Variations
%A Gupta, Ankita
%A Gunasekara, Chulaka
%A Wan, Hui
%A Ganhotra, Jatin
%A Joshi, Sachindra
%A Danilevsky, Marina
%Y Nouri, Elnaz
%Y Rastogi, Abhinav
%Y Spithourakis, Georgios
%Y Liu, Bing
%Y Chen, Yun-Nung
%Y Li, Yu
%Y Albalak, Alon
%Y Wakaki, Hiromi
%Y Papangelis, Alexandros
%S Proceedings of the 6th Workshop on NLP for Conversational AI (NLP4ConvAI 2024)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F gupta-etal-2024-evaluating
%X Dialogue summarization involves summarizing long conversations while preserving the most salient information. Real-life dialogues often involve naturally occurring variations (e.g., repetitions, hesitations). In this study, we systematically investigate the impact of such variations on state-of-the-art open dialogue summarization models whose details are publicly known (e.g., architectures, weights, and training corpora). To simulate real-life variations, we introduce two types of perturbations: utterance-level perturbations that modify individual utterances with errors and language variations, and dialogue-level perturbations that add non-informative exchanges (e.g., repetitions, greetings). We perform our analysis along three dimensions of robustness: consistency, saliency, and faithfulness, which aim to capture different aspects of performance of a summarization model. We find that both fine-tuned and instruction-tuned models are affected by input variations, with the latter being more susceptible, particularly to dialogue-level perturbations. We also validate our findings via human evaluation. Finally, we investigate whether the robustness of fine-tuned models can be improved by training them with a fraction of perturbed data. We find that this approach does not yield consistent performance gains, warranting further research. Overall, our work highlights robustness challenges in current open encoder-decoder summarization models and provides insights for future research.
%U https://aclanthology.org/2024.nlp4convai-1.4
%P 56-72
Markdown (Informal)
[Evaluating Robustness of Open Dialogue Summarization Models in the Presence of Naturally Occurring Variations](https://aclanthology.org/2024.nlp4convai-1.4) (Gupta et al., NLP4ConvAI-WS 2024)
ACL