OTTER: Improving Zero-Shot Classification via Optimal Transport
Popular zero-shot models suffer due to artifacts inherited from pretraining. A particularly detrimental artifact, caused by unbalanced web-scale pretraining data, is mismatched label distribution. Existing approaches that seek to repair the label distribution are not suitable in zero-shot settings,...
Saved in:
Published in | arXiv.org |
---|---|
Main Authors | , , , , |
Format | Paper |
Language | English |
Published |
Ithaca
Cornell University Library, arXiv.org
12.04.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Popular zero-shot models suffer due to artifacts inherited from pretraining. A particularly detrimental artifact, caused by unbalanced web-scale pretraining data, is mismatched label distribution. Existing approaches that seek to repair the label distribution are not suitable in zero-shot settings, as they have incompatible requirements such as access to labeled downstream task data or knowledge of the true label balance in the pretraining distribution. We sidestep these challenges and introduce a simple and lightweight approach to adjust pretrained model predictions via optimal transport. Our technique requires only an estimate of the label distribution of a downstream task. Theoretically, we characterize the improvement produced by our procedure under certain mild conditions and provide bounds on the error caused by misspecification. Empirically, we validate our method in a wide array of zero-shot image and text classification tasks, improving accuracy by 4.8% and 15.9% on average, and beating baselines like Prior Matching -- often by significant margins -- in 17 out of 21 datasets. |
---|---|
AbstractList | Popular zero-shot models suffer due to artifacts inherited from pretraining. A particularly detrimental artifact, caused by unbalanced web-scale pretraining data, is mismatched label distribution. Existing approaches that seek to repair the label distribution are not suitable in zero-shot settings, as they have incompatible requirements such as access to labeled downstream task data or knowledge of the true label balance in the pretraining distribution. We sidestep these challenges and introduce a simple and lightweight approach to adjust pretrained model predictions via optimal transport. Our technique requires only an estimate of the label distribution of a downstream task. Theoretically, we characterize the improvement produced by our procedure under certain mild conditions and provide bounds on the error caused by misspecification. Empirically, we validate our method in a wide array of zero-shot image and text classification tasks, improving accuracy by 4.8% and 15.9% on average, and beating baselines like Prior Matching -- often by significant margins -- in 17 out of 21 datasets. |
Author | Shin, Changho Cromp, Sonia Vishwakarma, Harit Zhao, Jitian Sala, Frederic |
Author_xml | – sequence: 1 givenname: Changho surname: Shin fullname: Shin, Changho – sequence: 2 givenname: Jitian surname: Zhao fullname: Zhao, Jitian – sequence: 3 givenname: Sonia surname: Cromp fullname: Cromp, Sonia – sequence: 4 givenname: Harit surname: Vishwakarma fullname: Vishwakarma, Harit – sequence: 5 givenname: Frederic surname: Sala fullname: Sala, Frederic |
BookMark | eNqNjLEKwjAUAIMoWLX_EHAupImx6uJQKjoJmsmlBEk1pc2reWm_3w5-gNMNd9yCTB04MyERFyJNdhvO5yRGrBljfJtxKUVEjlelituBXtrOw2Ddiz6Mh-T-hkDzRiPayj51sODoYDW9dsG2uqHKa4cd-LAis0o3aOIfl2R9KlR-TsbdpzcYyhp670ZVCib2LE1lJsV_1Rf8oDoD |
ContentType | Paper |
Copyright | 2024. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
Copyright_xml | – notice: 2024. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License. |
DBID | 8FE 8FG ABJCF ABUWG AFKRA AZQEC BENPR BGLVJ CCPQU DWQXO HCIFZ L6V M7S PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
DatabaseName | ProQuest SciTech Collection ProQuest Technology Collection Materials Science & Engineering Collection ProQuest Central (Alumni) ProQuest Central UK/Ireland ProQuest Central Essentials ProQuest Central Technology Collection ProQuest One Community College ProQuest Central SciTech Premium Collection (Proquest) (PQ_SDU_P3) ProQuest Engineering Collection ProQuest Engineering Database Publicly Available Content Database ProQuest One Academic Eastern Edition (DO NOT USE) ProQuest One Academic ProQuest One Academic UKI Edition ProQuest Central China Engineering Collection |
DatabaseTitle | Publicly Available Content Database Engineering Database Technology Collection ProQuest Central Essentials ProQuest One Academic Eastern Edition ProQuest Central (Alumni Edition) SciTech Premium Collection ProQuest One Community College ProQuest Technology Collection ProQuest SciTech Collection ProQuest Central China ProQuest Central ProQuest Engineering Collection ProQuest One Academic UKI Edition ProQuest Central Korea Materials Science & Engineering Collection ProQuest One Academic Engineering Collection |
DatabaseTitleList | Publicly Available Content Database |
Database_xml | – sequence: 1 dbid: 8FG name: ProQuest Technology Collection url: https://search.proquest.com/technologycollection1 sourceTypes: Aggregation Database |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Physics |
EISSN | 2331-8422 |
Genre | Working Paper/Pre-Print |
GroupedDBID | 8FE 8FG ABJCF ABUWG AFKRA ALMA_UNASSIGNED_HOLDINGS AZQEC BENPR BGLVJ CCPQU DWQXO FRJ HCIFZ L6V M7S M~E PIMPY PQEST PQQKQ PQUKI PRINS PTHSS |
ID | FETCH-proquest_journals_30390115753 |
IEDL.DBID | 8FG |
IngestDate | Tue Sep 24 21:20:28 EDT 2024 |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-proquest_journals_30390115753 |
OpenAccessLink | https://www.proquest.com/docview/3039011575/abstract/?pq-origsite=%requestingapplication% |
PQID | 3039011575 |
PQPubID | 2050157 |
ParticipantIDs | proquest_journals_3039011575 |
PublicationCentury | 2000 |
PublicationDate | 20240412 |
PublicationDateYYYYMMDD | 2024-04-12 |
PublicationDate_xml | – month: 04 year: 2024 text: 20240412 day: 12 |
PublicationDecade | 2020 |
PublicationPlace | Ithaca |
PublicationPlace_xml | – name: Ithaca |
PublicationTitle | arXiv.org |
PublicationYear | 2024 |
Publisher | Cornell University Library, arXiv.org |
Publisher_xml | – name: Cornell University Library, arXiv.org |
SSID | ssj0002672553 |
Score | 3.5237265 |
SecondaryResourceType | preprint |
Snippet | Popular zero-shot models suffer due to artifacts inherited from pretraining. A particularly detrimental artifact, caused by unbalanced web-scale pretraining... |
SourceID | proquest |
SourceType | Aggregation Database |
SubjectTerms | Image classification Labels |
Title | OTTER: Improving Zero-Shot Classification via Optimal Transport |
URI | https://www.proquest.com/docview/3039011575/abstract/ |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwfV3dS8MwED_miuDb_ELdHAF9DXX9ni8DpbUIm2MqDF9GkqYoTDu76qN_u3dhrQ_CHkMgIUdyd7_L_e4ALnUus9D3HZ4P3Yx7CjErkdK4E0ShRMQxFBHxnceTIH327uf-vAVpzYWhtMpaJxpFnRWKYuQ2qlpiSaJ3YQtJUQBV2aPVJ6f-UfTPummmsQPWgGriEWc8uWuiLU4Qou_s_lO4xookHbCmYqXLfWjpjwPYNcmXan0II4TV8eyaNQCfveiy4I-vRcVM00pK5zESZN9vgj3gK38XS9bUJT-CiyR-uk15vetic0PWi7_zuMfQRqivT4DJwNVUOPjKUcLzSU5aShH5Ua6o0Fx4Cr1tK51tn-7CHq5s8k4GTg_aVfmlz9GkVrJvpNUH6yaeTGc4Gv_Ev8qrgNg |
link.rule.ids | 786,790,12792,21416,33408,33779,43635,43840 |
linkProvider | ProQuest |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwY2BQSU1LSjE3NTXSTbM0TtE1SQb2WUGb0nSNzCzMk4A9DstEC9B-Z18_M49QE68I0wjogFsxdFklrEwEF9Qp-cmgMXJ9YFEL2iUJbF3YFxTqgm6NAs2uQq_QYGZgNTEGVp2gneJu7vAxFiMzc2CL2RijmAXXHW6CDKwBiQWpRUIMTKl5wgzs4CWXycUiDPbAzrRrkJUCvFuvEJValK8bnJFfogC-qhK0iAccbgplmYkK_sC8nZuYowA_jVyUQdnNNcTZQxdmazw0XRTHI3xhLMbAAuzgp0owKCSZGaeCjgs2MEpONDEFhU5qUlKihalFWjLoeDlzSQYZfCZJ4ZeWZ-D0CPH1iffx9POWZuAC2gJeeWJoJMPAUlJUmioLrFRLkuTAIQcAqZt9Ow |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=OTTER%3A+Improving+Zero-Shot+Classification+via+Optimal+Transport&rft.jtitle=arXiv.org&rft.au=Shin%2C+Changho&rft.au=Zhao%2C+Jitian&rft.au=Cromp%2C+Sonia&rft.au=Vishwakarma%2C+Harit&rft.date=2024-04-12&rft.pub=Cornell+University+Library%2C+arXiv.org&rft.eissn=2331-8422 |