@inproceedings{gupta-etal-2025-improving,
title = "Improving Model Evaluation using {SMART} Filtering of Benchmark Datasets",
author = "Gupta, Vipul and
Ross, Candace and
Pantoja, David and
Passonneau, Rebecca J. and
Ung, Megan and
Williams, Adina",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.235/",
doi = "10.18653/v1/2025.naacl-long.235",
pages = "4595--4615",
ISBN = "979-8-89176-189-6",
abstract = "One of the most challenging problems facing NLP today is evaluation. Some of the most pressing issues pertain to benchmark saturation, data contamination, and diversity in the quality of test examples. To address these concerns, we propose Selection Methodology for Accurate, Reduced, and Targeted (SMART) filtering, a novel approach to select a high-quality subset of examples from existing benchmark datasets by systematically removing less informative and lower quality examples. Our approach applies three filtering criteria, removing (i) easy examples, (ii) data-contaminated examples, and (iii) examples that are similar to each other based on distance in an embedding space. We demonstrate the effectiveness of SMART Filtering on three multiple choice QA datasets, where our methodology increases efficiency by reducing dataset size by 48{\%} on average, while increasing Pearson correlation with rankings from ChatBot Arena, a more open-ended human evaluation setting. Our method enables us to be more efficient, whether we are using SMART Filtering to make new benchmarks more challenging, or to revitalize older, human generated datasets, while still preserving the relative model rankings."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="gupta-etal-2025-improving">
<titleInfo>
<title>Improving Model Evaluation using SMART Filtering of Benchmark Datasets</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vipul</namePart>
<namePart type="family">Gupta</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Candace</namePart>
<namePart type="family">Ross</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Pantoja</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rebecca</namePart>
<namePart type="given">J</namePart>
<namePart type="family">Passonneau</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Megan</namePart>
<namePart type="family">Ung</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adina</namePart>
<namePart type="family">Williams</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>One of the most challenging problems facing NLP today is evaluation. Some of the most pressing issues pertain to benchmark saturation, data contamination, and diversity in the quality of test examples. To address these concerns, we propose Selection Methodology for Accurate, Reduced, and Targeted (SMART) filtering, a novel approach to select a high-quality subset of examples from existing benchmark datasets by systematically removing less informative and lower quality examples. Our approach applies three filtering criteria, removing (i) easy examples, (ii) data-contaminated examples, and (iii) examples that are similar to each other based on distance in an embedding space. We demonstrate the effectiveness of SMART Filtering on three multiple choice QA datasets, where our methodology increases efficiency by reducing dataset size by 48% on average, while increasing Pearson correlation with rankings from ChatBot Arena, a more open-ended human evaluation setting. Our method enables us to be more efficient, whether we are using SMART Filtering to make new benchmarks more challenging, or to revitalize older, human generated datasets, while still preserving the relative model rankings.</abstract>
<identifier type="citekey">gupta-etal-2025-improving</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.235</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.235/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>4595</start>
<end>4615</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving Model Evaluation using SMART Filtering of Benchmark Datasets
%A Gupta, Vipul
%A Ross, Candace
%A Pantoja, David
%A Passonneau, Rebecca J.
%A Ung, Megan
%A Williams, Adina
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F gupta-etal-2025-improving
%X One of the most challenging problems facing NLP today is evaluation. Some of the most pressing issues pertain to benchmark saturation, data contamination, and diversity in the quality of test examples. To address these concerns, we propose Selection Methodology for Accurate, Reduced, and Targeted (SMART) filtering, a novel approach to select a high-quality subset of examples from existing benchmark datasets by systematically removing less informative and lower quality examples. Our approach applies three filtering criteria, removing (i) easy examples, (ii) data-contaminated examples, and (iii) examples that are similar to each other based on distance in an embedding space. We demonstrate the effectiveness of SMART Filtering on three multiple choice QA datasets, where our methodology increases efficiency by reducing dataset size by 48% on average, while increasing Pearson correlation with rankings from ChatBot Arena, a more open-ended human evaluation setting. Our method enables us to be more efficient, whether we are using SMART Filtering to make new benchmarks more challenging, or to revitalize older, human generated datasets, while still preserving the relative model rankings.
%R 10.18653/v1/2025.naacl-long.235
%U https://aclanthology.org/2025.naacl-long.235/
%U https://doi.org/10.18653/v1/2025.naacl-long.235
%P 4595-4615
Markdown (Informal)
[Improving Model Evaluation using SMART Filtering of Benchmark Datasets](https://aclanthology.org/2025.naacl-long.235/) (Gupta et al., NAACL 2025)
ACL
- Vipul Gupta, Candace Ross, David Pantoja, Rebecca J. Passonneau, Megan Ung, and Adina Williams. 2025. Improving Model Evaluation using SMART Filtering of Benchmark Datasets. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 4595–4615, Albuquerque, New Mexico. Association for Computational Linguistics.