@inproceedings{chan-etal-2024-distribution-aware,
title = "Distribution Aware Metrics for Conditional Natural Language Generation",
author = "Chan, David M. and
Ni, Yiming and
Ross, David and
Vijayanarasimhan, Sudheendra and
Myers, Austin and
Canny, John",
editor = "Calzolari, Nicoletta and
Kan, Min-Yen and
Hoste, Veronique and
Lenci, Alessandro and
Sakti, Sakriani and
Xue, Nianwen",
booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
month = may,
year = "2024",
address = "Torino, Italia",
publisher = "ELRA and ICCL",
url = "https://aclanthology.org/2024.lrec-main.453",
pages = "5064--5095",
abstract = "Traditional automated metrics for evaluating conditional natural language generation rely on pairwise comparisons between a single generated text and the best-matching gold-standard reference. This method is effective when ground truth data diversity can be attributed to noise, however, it falls short when diversity in references holds valuable contextual information, as in visual description or summarization, as it does not evaluate the ability of a model to generate text matching the diversity of the ground truth samples. In this paper, we challenge the adequacy of existing metrics in such semantically diverse contexts and introduce a novel approach for evaluating conditional language generation models, leveraging a family of meta-metrics that build on existing pairwise distance functions. These meta-metrics assess not just single-samples, but distributions of reference and model-generated captions using small sample sets. We demonstrate our approach through a case study of visual description in the English language which reveals not only how current models prioritize single-description quality over diversity, but further sheds light on the impact of sampling methods and temperature settings on description quality and diversity.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chan-etal-2024-distribution-aware">
<titleInfo>
<title>Distribution Aware Metrics for Conditional Natural Language Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Chan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiming</namePart>
<namePart type="family">Ni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Ross</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sudheendra</namePart>
<namePart type="family">Vijayanarasimhan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Austin</namePart>
<namePart type="family">Myers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">John</namePart>
<namePart type="family">Canny</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicoletta</namePart>
<namePart type="family">Calzolari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min-Yen</namePart>
<namePart type="family">Kan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Veronique</namePart>
<namePart type="family">Hoste</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Lenci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nianwen</namePart>
<namePart type="family">Xue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA and ICCL</publisher>
<place>
<placeTerm type="text">Torino, Italia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Traditional automated metrics for evaluating conditional natural language generation rely on pairwise comparisons between a single generated text and the best-matching gold-standard reference. This method is effective when ground truth data diversity can be attributed to noise, however, it falls short when diversity in references holds valuable contextual information, as in visual description or summarization, as it does not evaluate the ability of a model to generate text matching the diversity of the ground truth samples. In this paper, we challenge the adequacy of existing metrics in such semantically diverse contexts and introduce a novel approach for evaluating conditional language generation models, leveraging a family of meta-metrics that build on existing pairwise distance functions. These meta-metrics assess not just single-samples, but distributions of reference and model-generated captions using small sample sets. We demonstrate our approach through a case study of visual description in the English language which reveals not only how current models prioritize single-description quality over diversity, but further sheds light on the impact of sampling methods and temperature settings on description quality and diversity.</abstract>
<identifier type="citekey">chan-etal-2024-distribution-aware</identifier>
<location>
<url>https://aclanthology.org/2024.lrec-main.453</url>
</location>
<part>
<date>2024-05</date>
<extent unit="page">
<start>5064</start>
<end>5095</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Distribution Aware Metrics for Conditional Natural Language Generation
%A Chan, David M.
%A Ni, Yiming
%A Ross, David
%A Vijayanarasimhan, Sudheendra
%A Myers, Austin
%A Canny, John
%Y Calzolari, Nicoletta
%Y Kan, Min-Yen
%Y Hoste, Veronique
%Y Lenci, Alessandro
%Y Sakti, Sakriani
%Y Xue, Nianwen
%S Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
%D 2024
%8 May
%I ELRA and ICCL
%C Torino, Italia
%F chan-etal-2024-distribution-aware
%X Traditional automated metrics for evaluating conditional natural language generation rely on pairwise comparisons between a single generated text and the best-matching gold-standard reference. This method is effective when ground truth data diversity can be attributed to noise, however, it falls short when diversity in references holds valuable contextual information, as in visual description or summarization, as it does not evaluate the ability of a model to generate text matching the diversity of the ground truth samples. In this paper, we challenge the adequacy of existing metrics in such semantically diverse contexts and introduce a novel approach for evaluating conditional language generation models, leveraging a family of meta-metrics that build on existing pairwise distance functions. These meta-metrics assess not just single-samples, but distributions of reference and model-generated captions using small sample sets. We demonstrate our approach through a case study of visual description in the English language which reveals not only how current models prioritize single-description quality over diversity, but further sheds light on the impact of sampling methods and temperature settings on description quality and diversity.
%U https://aclanthology.org/2024.lrec-main.453
%P 5064-5095
Markdown (Informal)
[Distribution Aware Metrics for Conditional Natural Language Generation](https://aclanthology.org/2024.lrec-main.453) (Chan et al., LREC-COLING 2024)
ACL
- David M. Chan, Yiming Ni, David Ross, Sudheendra Vijayanarasimhan, Austin Myers, and John Canny. 2024. Distribution Aware Metrics for Conditional Natural Language Generation. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 5064–5095, Torino, Italia. ELRA and ICCL.