{"id":"https://openalex.org/W7133325684","doi":"https://doi.org/10.48550/arxiv.2603.00857","title":"MultiPUFFIN: A Multimodal Domain-Constrained Foundation Model for Molecular Property Prediction of Small Molecules","display_name":"MultiPUFFIN: A Multimodal Domain-Constrained Foundation Model for Molecular Property Prediction of Small Molecules","publication_year":2026,"publication_date":"2026-03-01","ids":{"openalex":"https://openalex.org/W7133325684","doi":"https://doi.org/10.48550/arxiv.2603.00857"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.00857","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00857","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.00857","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5121156890","display_name":"Idelfonso B. R. Nogueira","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Nogueira, Idelfonso B. R.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127920358","display_name":"Carine M. Rebelloa","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rebelloa, Carine M.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051279714","display_name":"M. Enis Leblebici","orcid":"https://orcid.org/0000-0003-4599-9412"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Leblebici, Mumin Enis","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5004611294","display_name":"Erick Giovani Sperandio Nascimento","orcid":"https://orcid.org/0000-0003-2219-0290"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nascimento, Erick Giovani Sperandio","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5121156890"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.8842999935150146,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.8842999935150146,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10211","display_name":"Computational Drug Discovery Methods","score":0.07850000262260437,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10798","display_name":"Crystallography and molecular interactions","score":0.007400000002235174,"subfield":{"id":"https://openalex.org/subfields/1606","display_name":"Physical and Theoretical Chemistry"},"field":{"id":"https://openalex.org/fields/16","display_name":"Chemistry"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/chemical-space","display_name":"Chemical space","score":0.6998999714851379},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5694000124931335},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.5435000061988831},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5393999814987183},{"id":"https://openalex.org/keywords/property","display_name":"Property (philosophy)","score":0.5260000228881836},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.5250999927520752},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.5113999843597412},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4620000123977661},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4577000141143799}],"concepts":[{"id":"https://openalex.org/C99726746","wikidata":"https://www.wikidata.org/wiki/Q906396","display_name":"Chemical space","level":3,"score":0.6998999714851379},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5694000124931335},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.5435000061988831},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5393999814987183},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5354999899864197},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.5260000228881836},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.5250999927520752},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.5113999843597412},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4620000123977661},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4577000141143799},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.43939998745918274},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4343999922275543},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3806000053882599},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.3400999903678894},{"id":"https://openalex.org/C101738243","wikidata":"https://www.wikidata.org/wiki/Q786435","display_name":"Autoencoder","level":3,"score":0.33809998631477356},{"id":"https://openalex.org/C31555180","wikidata":"https://www.wikidata.org/wiki/Q3523867","display_name":"Material properties","level":2,"score":0.33320000767707825},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.32280001044273376},{"id":"https://openalex.org/C148043351","wikidata":"https://www.wikidata.org/wiki/Q4456944","display_name":"Current (fluid)","level":2,"score":0.2973000109195709},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27709999680519104},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.27469998598098755},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.2694000005722046},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26570001244544983},{"id":"https://openalex.org/C2781311116","wikidata":"https://www.wikidata.org/wiki/Q83306","display_name":"Group (periodic table)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.25760000944137573},{"id":"https://openalex.org/C2994228675","wikidata":"https://www.wikidata.org/wiki/Q512599","display_name":"Property value","level":3,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.00857","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00857","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.00857","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00857","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Predicting":[0],"physicochemical":[1],"properties":[2,30,101,153],"across":[3,150],"chemical":[4,9],"space":[5],"is":[6,104],"vital":[7],"for":[8,164],"engineering,":[10],"drug":[11],"discovery,":[12],"and":[13,31,54,64,78,88,182,188,200],"materials":[14],"science.":[15],"Current":[16],"molecular":[17],"foundation":[18,40],"models":[19],"lack":[20],"thermodynamic":[21,86,175,203],"consistency,":[22],"while":[23],"domain-informed":[24,183,217],"approaches":[25],"are":[26,161],"limited":[27],"to":[28,84,138,173,192],"single":[29],"small":[32],"datasets.":[33],"We":[34],"introduce":[35],"MultiPUFFIN,":[36],"a":[37,90,107,123,129],"domain-constrained":[38],"multimodal":[39,180],"model":[41],"addressing":[42],"both":[43],"limitations":[44],"simultaneously.":[45,102],"MultiPUFFIN":[46,97,121,145,196],"features:":[47],"(i)":[48],"an":[49],"encoder":[50],"fusing":[51],"SMILES,":[52],"graphs,":[53],"3D":[55],"geometries":[56],"via":[57],"gated":[58],"cross-modal":[59],"attention,":[60],"alongside":[61],"experimental":[62],"condition":[63],"descriptor":[65],"encoders;":[66],"(ii)":[67],"prediction":[68,218],"heads":[69],"embedding":[70],"established":[71],"correlations":[72],"(e.g.,":[73],"Wagner,":[74],"Andrade,":[75],"van't":[76],"Hoff,":[77],"Shomate":[79],"equations)":[80],"as":[81],"inductive":[82],"biases":[83,184],"ensure":[85],"consistency;":[87],"(iii)":[89],"two-stage":[91],"multi-task":[92],"training":[93,158],"strategy.Extending":[94],"prior":[95],"frameworks,":[96],"predicts":[98],"nine":[99,152],"thermophysical":[100],"It":[103],"trained":[105],"on":[106,128,141],"multi-source":[108],"dataset":[109],"of":[110,134,215],"37,968":[111],"unique":[112],"molecules":[113],"(40,904":[114],"rows).":[115],"With":[116],"roughly":[117],"35":[118],"million":[119,143],"parameters,":[120],"achieves":[122],"mean":[124],"$R^2":[125],"=":[126],"0.716$":[127],"challenging":[130],"scaffold-split":[131],"test":[132],"set":[133],"8,877":[135],"molecules.":[136,159],"Compared":[137],"ChemBERTa-2":[139,168],"(pre-trained":[140],"77":[142],"molecules),":[144],"outperforms":[146],"the":[147,170,212],"fine-tuned":[148],"baseline":[149],"all":[151],"despite":[154],"using":[155],"2000x":[156],"fewer":[157],"Advantages":[160],"strikingly":[162],"apparent":[163],"temperature-dependent":[165],"properties,":[166],"where":[167],"lacks":[169],"architectural":[171],"capacity":[172],"incorporate":[174],"conditions.These":[176],"results":[177],"demonstrate":[178],"that":[179],"encoding":[181],"substantially":[185],"reduce":[186],"data":[187],"compute":[189],"requirements":[190],"compared":[191],"brute-force":[193],"pre-training.":[194],"Furthermore,":[195],"handles":[197],"missing":[198],"modalities":[199],"recovers":[201],"meaningful":[202],"parameters":[204],"without":[205],"explicit":[206],"supervision.":[207],"Systematic":[208],"ablation":[209],"studies":[210],"confirm":[211],"property-specific":[213],"benefits":[214],"these":[216],"heads.":[219]},"counts_by_year":[],"updated_date":"2026-03-04T07:09:34.246503","created_date":"2026-03-04T00:00:00"}