{"id":"https://openalex.org/W7162667420","doi":"https://doi.org/10.48550/arxiv.2605.28287","title":"AtomComposer: Discovering Chemical Space from First Principles with Reinforcement Learning","display_name":"AtomComposer: Discovering Chemical Space from First Principles with Reinforcement Learning","publication_year":2026,"publication_date":"2026-05-27","ids":{"openalex":"https://openalex.org/W7162667420","doi":"https://doi.org/10.48550/arxiv.2605.28287"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.28287","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28287","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.28287","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137281918","display_name":"Bjarke Hastrup","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hastrup, Bjarke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137307085","display_name":"Francois Cornet","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cornet, Francois","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137228668","display_name":"Tejs Vegge","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vegge, Tejs","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5023121476","display_name":"Arghya Bhowmik","orcid":"https://orcid.org/0000-0003-3198-5116"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bhowmik, Arghya","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"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":0.9930999875068665,"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":0.9930999875068665,"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":0.0017000000225380063,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0003000000142492354,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.7817000150680542},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7276999950408936},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.6035000085830688},{"id":"https://openalex.org/keywords/chemical-space","display_name":"Chemical space","score":0.4595000147819519},{"id":"https://openalex.org/keywords/limit","display_name":"Limit (mathematics)","score":0.42669999599456787},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.42010000348091125},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.40560001134872437},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.3422999978065491}],"concepts":[{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.7817000150680542},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7276999950408936},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6492999792098999},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.629800021648407},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6035000085830688},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4726000130176544},{"id":"https://openalex.org/C99726746","wikidata":"https://www.wikidata.org/wiki/Q906396","display_name":"Chemical space","level":3,"score":0.4595000147819519},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.42669999599456787},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.42010000348091125},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.40560001134872437},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3422999978065491},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33799999952316284},{"id":"https://openalex.org/C48164120","wikidata":"https://www.wikidata.org/wiki/Q4491893","display_name":"Concept learning","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C2986087404","wikidata":"https://www.wikidata.org/wiki/Q15946010","display_name":"Online learning","level":2,"score":0.29840001463890076},{"id":"https://openalex.org/C74072328","wikidata":"https://www.wikidata.org/wiki/Q1142726","display_name":"Intelligent agent","level":2,"score":0.2809000015258789},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.27070000767707825},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.28287","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28287","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.28287","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.28287","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":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.4556110203266144,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Discovering":[0],"novel":[1,29],"stable":[2],"molecules":[3],"without":[4,48],"training":[5,93],"data":[6],"remains":[7],"a":[8,35,58,85,91,97,146],"grand":[9],"scientific":[10],"challenge.":[11],"Current":[12],"molecular":[13],"generative":[14],"models":[15],"are":[16],"trained":[17,72,131],"on":[18,122],"large,":[19],"pre-curated":[20],"datasets,":[21],"which":[22],"introduce":[23],"biases":[24],"and":[25,70,106],"limit":[26],"exploration":[27,152],"of":[28,42,117,141,153],"chemistry.":[30],"In":[31],"contrast,":[32],"we":[33,55,89],"propose":[34],"new":[36],"paradigm:":[37],"autonomous,":[38],"generalized":[39],"agents":[40],"capable":[41],"mapping":[43],"vast,":[44],"unknown":[45],"chemical":[46,87,154],"spaces":[47],"any":[49],"pretraining.":[50],"For":[51],"the":[52,139],"first":[53],"time,":[54],"present":[56],"AtomComposer,":[57],"self-guided":[59],"agent":[60,110],"that":[61,81,95],"autonomously":[62],"constructs":[63],"valid":[64,120],"3D":[65],"isomers":[66,121],"under":[67],"stoichiometric":[68],"constraints":[69],"is":[71],"exclusively":[73],"online":[74,142],"using":[75],"reinforcement":[76,143],"learning.":[77],"Unlike":[78],"existing":[79,127],"approaches":[80],"generally":[82],"overfit":[83],"to":[84,114],"specific":[86],"formula,":[88],"establish":[90],"multi-composition":[92],"scheme":[94],"enables":[96],"broad":[98],"generalization":[99],"across":[100],"diverse":[101],"chemistry,":[102],"guided":[103],"by":[104],"energy-":[105],"validity-based":[107],"rewards.":[108,135],"Our":[109],"can":[111],"discover":[112],"up":[113],"an":[115],"order":[116],"magnitude":[118],"more":[119],"unseen":[123],"test":[124],"formulas":[125],"than":[126],"single-composition":[128],"reinforcement-learning":[129],"baselines":[130],"with":[132],"per-step":[133],"energy":[134],"These":[136],"results":[137],"fulfill":[138],"promise":[140],"learning":[144],"as":[145],"powerful":[147],"paradigm":[148],"for":[149],"scalable,":[150],"from-scratch":[151],"configuration":[155],"space.":[156]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-29T00:00:00"}