Preference Optimization for Molecular Language Models

Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not \emph{a priori} encode certain preferences a chemist may desire. We investigate the use of fine-tuning using Direct Preference Optimization to better align generated molecules w...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2023-10
Main Authors: Park, Ryan, Theisen, Ryan, Sahni, Navriti, Patek, Marcel, Cichońska, Anna, Rahman, Rayees
Format: Article
Language:English
Subjects:
Chemists
Optimization
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Park, Ryan
Theisen, Ryan
Sahni, Navriti
Patek, Marcel
Cichońska, Anna
Rahman, Rayees
description Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not \emph{a priori} encode certain preferences a chemist may desire. We investigate the use of fine-tuning using Direct Preference Optimization to better align generated molecules with chemist preferences. Our findings suggest that this approach is simple, efficient, and highly effective.
format article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2879448254</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2879448254</sourcerecordid><originalsourceid>FETCH-proquest_journals_28794482543</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwDShKTUstSs1LTlXwLyjJzM2sSizJzM9TSMsvUvDNz0lNLs1JLFLwScxLL01MTwUKpaTmFPMwsKYl5hSn8kJpbgZlN9cQZw_dgqL8wtLU4pL4rPzSojygVLyRhbmliYmFkamJMXGqAK-yNT0</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2879448254</pqid></control><display><type>article</type><title>Preference Optimization for Molecular Language Models</title><source>Publicly Available Content Database</source><creator>Park, Ryan ; Theisen, Ryan ; Sahni, Navriti ; Patek, Marcel ; Cichońska, Anna ; Rahman, Rayees</creator><creatorcontrib>Park, Ryan ; Theisen, Ryan ; Sahni, Navriti ; Patek, Marcel ; Cichońska, Anna ; Rahman, Rayees</creatorcontrib><description>Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not \emph{a priori} encode certain preferences a chemist may desire. We investigate the use of fine-tuning using Direct Preference Optimization to better align generated molecules with chemist preferences. Our findings suggest that this approach is simple, efficient, and highly effective.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Chemists ; Optimization</subject><ispartof>arXiv.org, 2023-10</ispartof><rights>2023. 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.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://www.proquest.com/docview/2879448254?pq-origsite=primo$$EHTML$$P50$$Gproquest$$Hfree_for_read</linktohtml><link.rule.ids>777,781,25734,36993,44571</link.rule.ids></links><search><creatorcontrib>Park, Ryan</creatorcontrib><creatorcontrib>Theisen, Ryan</creatorcontrib><creatorcontrib>Sahni, Navriti</creatorcontrib><creatorcontrib>Patek, Marcel</creatorcontrib><creatorcontrib>Cichońska, Anna</creatorcontrib><creatorcontrib>Rahman, Rayees</creatorcontrib><title>Preference Optimization for Molecular Language Models</title><title>arXiv.org</title><description>Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not \emph{a priori} encode certain preferences a chemist may desire. We investigate the use of fine-tuning using Direct Preference Optimization to better align generated molecules with chemist preferences. Our findings suggest that this approach is simple, efficient, and highly effective.</description><subject>Chemists</subject><subject>Optimization</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>PIMPY</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwDShKTUstSs1LTlXwLyjJzM2sSizJzM9TSMsvUvDNz0lNLs1JLFLwScxLL01MTwUKpaTmFPMwsKYl5hSn8kJpbgZlN9cQZw_dgqL8wtLU4pL4rPzSojygVLyRhbmliYmFkamJMXGqAK-yNT0</recordid><startdate>20231018</startdate><enddate>20231018</enddate><creator>Park, Ryan</creator><creator>Theisen, Ryan</creator><creator>Sahni, Navriti</creator><creator>Patek, Marcel</creator><creator>Cichońska, Anna</creator><creator>Rahman, Rayees</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20231018</creationdate><title>Preference Optimization for Molecular Language Models</title><author>Park, Ryan ; Theisen, Ryan ; Sahni, Navriti ; Patek, Marcel ; Cichońska, Anna ; Rahman, Rayees</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_28794482543</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Chemists</topic><topic>Optimization</topic><toplevel>online_resources</toplevel><creatorcontrib>Park, Ryan</creatorcontrib><creatorcontrib>Theisen, Ryan</creatorcontrib><creatorcontrib>Sahni, Navriti</creatorcontrib><creatorcontrib>Patek, Marcel</creatorcontrib><creatorcontrib>Cichońska, Anna</creatorcontrib><creatorcontrib>Rahman, Rayees</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni)</collection><collection>ProQuest Central</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Databases</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Publicly Available Content Database</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Park, Ryan</au><au>Theisen, Ryan</au><au>Sahni, Navriti</au><au>Patek, Marcel</au><au>Cichońska, Anna</au><au>Rahman, Rayees</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Preference Optimization for Molecular Language Models</atitle><jtitle>arXiv.org</jtitle><date>2023-10-18</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not \emph{a priori} encode certain preferences a chemist may desire. We investigate the use of fine-tuning using Direct Preference Optimization to better align generated molecules with chemist preferences. Our findings suggest that this approach is simple, efficient, and highly effective.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-10
issn 2331-8422
language eng
recordid cdi_proquest_journals_2879448254
source Publicly Available Content Database
subjects Chemists
Optimization
title Preference Optimization for Molecular Language Models
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-19T20%3A38%3A13IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Preference%20Optimization%20for%20Molecular%20Language%20Models&rft.jtitle=arXiv.org&rft.au=Park,%20Ryan&rft.date=2023-10-18&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2879448254%3C/proquest%3E%3Cgrp_id%3Ecdi_FETCH-proquest_journals_28794482543%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=2879448254&rft_id=info:pmid/&rfr_iscdi=true