@inproceedings{awadalla-etal-2022-exploring,
title = "Exploring The Landscape of Distributional Robustness for Question Answering Models",
author = "Awadalla, Anas and
Wortsman, Mitchell and
Ilharco, Gabriel and
Min, Sewon and
Magnusson, Ian and
Hajishirzi, Hannaneh and
Schmidt, Ludwig",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.441",
doi = "10.18653/v1/2022.findings-emnlp.441",
pages = "5971--5987",
abstract = "We conduct a large empirical evaluation to investigate the landscape of distributional robustness in question answering. Our investigation spans over 350 models and 16 question answering datasets, including a diverse set of architectures, model sizes, and adaptation methods (e.g., fine-tuning, adapter tuning, in-context learning, etc.). We find that, in many cases, model variations do not affect robustness and in-distribution performance alone determines out-of-distribution performance. Moreover, our findings indicate thati) zero-shot and in-context learning methods are more robust to distribution shifts than fully fine-tuned models;ii) few-shot prompt fine-tuned models exhibit better robustness than few-shot fine-tuned span prediction models;iii) parameter-efficient and robustness enhancing training methods provide no significant robustness improvements. In addition, we publicly release all evaluations to encourage researchers to further analyze robustness trends for question answering models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="awadalla-etal-2022-exploring">
<titleInfo>
<title>Exploring The Landscape of Distributional Robustness for Question Answering Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anas</namePart>
<namePart type="family">Awadalla</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mitchell</namePart>
<namePart type="family">Wortsman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gabriel</namePart>
<namePart type="family">Ilharco</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sewon</namePart>
<namePart type="family">Min</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ian</namePart>
<namePart type="family">Magnusson</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hannaneh</namePart>
<namePart type="family">Hajishirzi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ludwig</namePart>
<namePart type="family">Schmidt</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We conduct a large empirical evaluation to investigate the landscape of distributional robustness in question answering. Our investigation spans over 350 models and 16 question answering datasets, including a diverse set of architectures, model sizes, and adaptation methods (e.g., fine-tuning, adapter tuning, in-context learning, etc.). We find that, in many cases, model variations do not affect robustness and in-distribution performance alone determines out-of-distribution performance. Moreover, our findings indicate thati) zero-shot and in-context learning methods are more robust to distribution shifts than fully fine-tuned models;ii) few-shot prompt fine-tuned models exhibit better robustness than few-shot fine-tuned span prediction models;iii) parameter-efficient and robustness enhancing training methods provide no significant robustness improvements. In addition, we publicly release all evaluations to encourage researchers to further analyze robustness trends for question answering models.</abstract>
<identifier type="citekey">awadalla-etal-2022-exploring</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.441</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.441</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>5971</start>
<end>5987</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Exploring The Landscape of Distributional Robustness for Question Answering Models
%A Awadalla, Anas
%A Wortsman, Mitchell
%A Ilharco, Gabriel
%A Min, Sewon
%A Magnusson, Ian
%A Hajishirzi, Hannaneh
%A Schmidt, Ludwig
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F awadalla-etal-2022-exploring
%X We conduct a large empirical evaluation to investigate the landscape of distributional robustness in question answering. Our investigation spans over 350 models and 16 question answering datasets, including a diverse set of architectures, model sizes, and adaptation methods (e.g., fine-tuning, adapter tuning, in-context learning, etc.). We find that, in many cases, model variations do not affect robustness and in-distribution performance alone determines out-of-distribution performance. Moreover, our findings indicate thati) zero-shot and in-context learning methods are more robust to distribution shifts than fully fine-tuned models;ii) few-shot prompt fine-tuned models exhibit better robustness than few-shot fine-tuned span prediction models;iii) parameter-efficient and robustness enhancing training methods provide no significant robustness improvements. In addition, we publicly release all evaluations to encourage researchers to further analyze robustness trends for question answering models.
%R 10.18653/v1/2022.findings-emnlp.441
%U https://aclanthology.org/2022.findings-emnlp.441
%U https://doi.org/10.18653/v1/2022.findings-emnlp.441
%P 5971-5987
Markdown (Informal)
[Exploring The Landscape of Distributional Robustness for Question Answering Models](https://aclanthology.org/2022.findings-emnlp.441) (Awadalla et al., Findings 2022)
ACL