{"id":"https://openalex.org/W7154044755","doi":"https://doi.org/10.48550/arxiv.2604.09130","title":"EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers","display_name":"EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers","publication_year":2026,"publication_date":"2026-04-10","ids":{"openalex":"https://openalex.org/W7154044755","doi":"https://doi.org/10.48550/arxiv.2604.09130"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.09130","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09130","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2604.09130","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058593712","display_name":"Yi-Lun Liao","orcid":"https://orcid.org/0000-0002-5299-6749"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Liao, Yi-Lun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002799648","display_name":"Alexander J. Hoffman","orcid":"https://orcid.org/0000-0002-1337-9297"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hoffman, Alexander J.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004258503","display_name":"Sabrina C. Shen","orcid":"https://orcid.org/0000-0002-6562-3829"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Sabrina C.","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083981921","display_name":"Alexandre Duval","orcid":"https://orcid.org/0000-0001-9416-3270"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Duval, Alexandre","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051229816","display_name":"Sam Walton Norwood","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Norwood, Sam Walton","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5062301881","display_name":"Tess Smidt","orcid":"https://orcid.org/0000-0001-5581-5344"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Smidt, Tess","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5058593712"],"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":0.9793999791145325,"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":0.9793999791145325,"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.003100000089034438,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T11804","display_name":"Quantum many-body systems","score":0.003100000089034438,"subfield":{"id":"https://openalex.org/subfields/3107","display_name":"Atomic and Molecular Physics, and Optics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.4724000096321106},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.4260999858379364},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.42179998755455017},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.3903999924659729},{"id":"https://openalex.org/keywords/feed-forward","display_name":"Feed forward","score":0.38839998841285706},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.3776000142097473}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.691100001335144},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.5358999967575073},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.4724000096321106},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.4260999858379364},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.42179998755455017},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3903999924659729},{"id":"https://openalex.org/C38858127","wikidata":"https://www.wikidata.org/wiki/Q5441228","display_name":"Feed forward","level":2,"score":0.38839998841285706},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3776000142097473},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34380000829696655},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.3239000141620636},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.32330000400543213},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3059999942779541},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.298799991607666},{"id":"https://openalex.org/C2993807640","wikidata":"https://www.wikidata.org/wiki/Q103709453","display_name":"Attention network","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C171036898","wikidata":"https://www.wikidata.org/wiki/Q256355","display_name":"Equivariant map","level":2,"score":0.27709999680519104},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2712000012397766},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.25850000977516174}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.09130","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09130","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.09130","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.09130","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"score":0.8357704281806946,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"As":[0],"$SE(3)$-equivariant":[1,40],"graph":[2,41],"neural":[3],"networks":[4],"mature":[5],"as":[6],"a":[7,23],"core":[8],"tool":[9],"for":[10,26,106],"3D":[11],"atomistic":[12],"modeling,":[13],"improving":[14],"their":[15],"efficiency,":[16,50],"expressivity,":[17,51],"and":[18,52,77,91,110,126,146,171],"physical":[19],"consistency":[20],"has":[21],"become":[22],"central":[24],"challenge":[25],"large-scale":[27],"applications.":[28],"In":[29],"this":[30],"work,":[31],"we":[32,57,65,74,98],"introduce":[33,75],"EquiformerV3,":[34],"the":[35,39,59,67,117,157],"third":[36],"generation":[37],"of":[38,119,132,149,160],"attention":[42,92,128],"Transformer,":[43],"designed":[44],"to":[45,80,102,111,141],"advance":[46],"all":[47],"three":[48,61],"dimensions:":[49],"generality.":[53],"Building":[54],"on":[55,168],"EquiformerV2,":[56,81],"have":[58],"following":[60],"key":[62],"advances.":[63],"First,":[64],"optimize":[66],"software":[68],"implementation,":[69],"achieving":[70],"$1.75\\times$":[71],"speedup.":[72],"Second,":[73],"simple":[76],"effective":[78],"modifications":[79],"including":[82],"equivariant":[83],"merged":[84],"layer":[85],"normalization,":[86],"improved":[87],"feedforward":[88],"network":[89],"hyper-parameters,":[90],"with":[93,156],"smooth":[94],"radius":[95],"cutoff.":[96],"Third,":[97],"propose":[99],"SwiGLU-$S^2$":[100,124],"activations":[101,125],"incorporate":[103],"many-body":[104],"interactions":[105],"better":[107],"theoretical":[108],"expressivity":[109],"preserve":[112],"strict":[113],"equivariance":[114],"while":[115],"reducing":[116],"complexity":[118],"sampling":[120],"$S^2$":[121],"grids.":[122],"Together,":[123],"smooth-cutoff":[127],"enable":[129],"accurate":[130],"modeling":[131],"smoothly":[133],"varying":[134],"potential":[135],"energy":[136],"surfaces":[137],"(PES),":[138],"generalizing":[139],"EquiformerV3":[140,154],"tasks":[142],"requiring":[143],"energy-conserving":[144],"simulations":[145],"higher-order":[147],"derivatives":[148],"PES.":[150],"With":[151],"these":[152],"improvements,":[153],"trained":[155],"auxiliary":[158],"task":[159],"denoising":[161],"non-equilibrium":[162],"structures":[163],"(DeNS)":[164],"achieves":[165],"state-of-the-art":[166],"results":[167],"OC20,":[169],"OMat24,":[170],"Matbench":[172],"Discovery.":[173]},"counts_by_year":[],"updated_date":"2026-04-14T06:08:25.285971","created_date":"2026-04-14T00:00:00"}