{"id":"https://openalex.org/W7212084319","doi":"https://doi.org/10.26434/chemrxiv.15008550/v1","title":"Contrastive Reward Based Reinforcement Learning for Goal Directed 3D Molecular Design","display_name":"Contrastive Reward Based Reinforcement Learning for Goal Directed 3D Molecular Design","publication_year":2026,"publication_date":"2026-09-09","ids":{"openalex":"https://openalex.org/W7212084319","doi":"https://doi.org/10.26434/chemrxiv.15008550/v1"},"language":"en","primary_location":{"id":"doi:10.26434/chemrxiv.15008550/v1","doi":"https://doi.org/10.26434/chemrxiv.15008550/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15008550/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008550/v1","source":{"id":"https://openalex.org/S4393918830","display_name":"ChemRxiv","issn_l":"2573-2293","issn":["2573-2293"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008550/v1","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133550341","display_name":"Bohao Li","orcid":"https://orcid.org/0009-0004-0627-806X","observed_orcids":["https://orcid.org/0009-0004-0627-806X"]},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]},{"id":"https://openalex.org/I4405252760","display_name":"Guangzhou Laboratory","ror":"https://ror.org/03ybmxt82","country_code":"CN","type":"government","lineage":["https://openalex.org/I4405252760"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bohao Li","raw_affiliation_strings":["Guangzhou National Laboratory","School of Computer Science and Engineering","Sun Yat-Sen University"],"raw_orcid":"https://orcid.org/0009-0004-0627-806X","affiliations":[{"raw_affiliation_string":"Guangzhou National Laboratory","institution_ids":["https://openalex.org/I4405252760"]},{"raw_affiliation_string":"School of Computer Science and Engineering","institution_ids":[]},{"raw_affiliation_string":"Sun Yat-Sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023539493","display_name":"Yuedong Yang","orcid":"https://orcid.org/0000-0002-6782-2813","observed_orcids":["https://orcid.org/0000-0002-6782-2813"]},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuedong Yang","raw_affiliation_strings":["School of Computer Science and Engineering","Sun Yat-Sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computer Science and Engineering","institution_ids":[]},{"raw_affiliation_string":"Sun Yat-Sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134442907","display_name":"Mingyuan Xu","orcid":"https://orcid.org/0000-0003-0463-4249","observed_orcids":["https://orcid.org/0000-0003-0463-4249"]},"institutions":[{"id":"https://openalex.org/I4405252760","display_name":"Guangzhou Laboratory","ror":"https://ror.org/03ybmxt82","country_code":"CN","type":"government","lineage":["https://openalex.org/I4405252760"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mingyuan Xu","raw_affiliation_strings":["Guangzhou National Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Guangzhou National Laboratory","institution_ids":["https://openalex.org/I4405252760"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135259207","display_name":"Hongming Chen","orcid":"https://orcid.org/0000-0002-8065-8333","observed_orcids":["https://orcid.org/0000-0002-8065-8333"]},"institutions":[{"id":"https://openalex.org/I4405252760","display_name":"Guangzhou Laboratory","ror":"https://ror.org/03ybmxt82","country_code":"CN","type":"government","lineage":["https://openalex.org/I4405252760"]},{"id":"https://openalex.org/I92039509","display_name":"Guangzhou Medical University","ror":"https://ror.org/00zat6v61","country_code":"CN","type":"education","lineage":["https://openalex.org/I92039509"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hongming Chen","raw_affiliation_strings":["Guangzhou Medical University","Guangzhou National Laboratory","School of Pharmaceutical Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Guangzhou Medical University","institution_ids":["https://openalex.org/I92039509"]},{"raw_affiliation_string":"Guangzhou National Laboratory","institution_ids":["https://openalex.org/I4405252760"]},{"raw_affiliation_string":"School of Pharmaceutical Sciences","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10211","display_name":"Computational Drug Discovery Methods","score":0.9854000210762024,"subfield":{"id":"https://openalex.org/subfields/3002","display_name":"Drug Discovery"},"field":{"id":"https://openalex.org/fields/30","display_name":"Pharmacology, Toxicology and Pharmaceutics"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T10211","display_name":"Computational Drug Discovery Methods","score":0.9854000210762024,"subfield":{"id":"https://openalex.org/subfields/3002","display_name":"Drug Discovery"},"field":{"id":"https://openalex.org/fields/30","display_name":"Pharmacology, Toxicology and Pharmaceutics"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.005100000184029341,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0008999999845400453,"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/contrastive-learning","display_name":"contrastive learning","score":0.8669999837875366},{"id":"https://openalex.org/keywords/molecular-generative-models","display_name":"molecular generative models","score":0.5540000200271606},{"id":"https://openalex.org/keywords/shape-optimization","display_name":"shape optimization","score":0.5870000123977661},{"id":"https://openalex.org/keywords/anchor-loss","display_name":"anchor loss","score":0.5509999990463257}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7547000050544739},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7164999842643738},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.6514999866485596},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.5842000246047974},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5601999759674072},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5073999762535095},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.4936000108718872},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.4909999966621399},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4560999870300293},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.4442000091075897},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.42239999771118164},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.40639999508857727},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.3662000000476837},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.32580000162124634},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.32519999146461487},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.31119999289512634},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.3102000057697296},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.29429998993873596},{"id":"https://openalex.org/C2778445095","wikidata":"https://www.wikidata.org/wiki/Q18354077","display_name":"Sample complexity","level":2,"score":0.2793999910354614},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.27250000834465027},{"id":"https://openalex.org/C55974624","wikidata":"https://www.wikidata.org/wiki/Q1188504","display_name":"Exponential family","level":2,"score":0.2700999975204468},{"id":"https://openalex.org/C40506919","wikidata":"https://www.wikidata.org/wiki/Q7452469","display_name":"Sequence learning","level":2,"score":0.2671000063419342},{"id":"https://openalex.org/C49781872","wikidata":"https://www.wikidata.org/wiki/Q1045555","display_name":"Maximum likelihood","level":2,"score":0.257999986410141},{"id":"https://openalex.org/C194051981","wikidata":"https://www.wikidata.org/wiki/Q1337691","display_name":"Economic shortage","level":3,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.26434/chemrxiv.15008550/v1","doi":"https://doi.org/10.26434/chemrxiv.15008550/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15008550/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008550/v1","source":{"id":"https://openalex.org/S4393918830","display_name":"ChemRxiv","issn_l":"2573-2293","issn":["2573-2293"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.26434/chemrxiv.15008550/v1","doi":"https://doi.org/10.26434/chemrxiv.15008550/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15008550/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008550/v1","source":{"id":"https://openalex.org/S4393918830","display_name":"ChemRxiv","issn_l":"2573-2293","issn":["2573-2293"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G431056085","display_name":null,"funder_award_id":"2026ZD01999600","funder_id":"https://openalex.org/F4320329860","funder_display_name":"National Science and Technology Major Project"}],"funders":[{"id":"https://openalex.org/F4320329860","display_name":"National Science and Technology Major Project","ror":null}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7212084319.pdf","grobid_xml":"https://content.openalex.org/works/W7212084319.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W2736601468","https://openalex.org/W4391631327","https://openalex.org/W4416627425","https://openalex.org/W7077449243","https://openalex.org/W7131386686"],"related_works":["https://openalex.org/W4416627425","https://openalex.org/W6910616346","https://openalex.org/W4413969676","https://openalex.org/W7113038875","https://openalex.org/W2947643486","https://openalex.org/W7101471270","https://openalex.org/W3106621283","https://openalex.org/W7170081715","https://openalex.org/W7134068474","https://openalex.org/W4414827321"],"abstract_inverted_index":{"Although":[0],"three-dimensional":[1,126],"flow-matching":[2,20,76,88,140],"models":[3,21,141],"have":[4],"become":[5],"powerful":[6],"tools":[7],"for":[8,19,32,70,86,164],"efficient":[9],"sampling":[10],"of":[11,27,73,115,139],"high-quality":[12],"3D":[13,131],"molecular":[14,34,75,132,151,166],"structures,":[15],"reward-based":[16],"reinforcement":[17,67],"learning":[18,68],"is":[22],"complicated":[23],"by":[24,48],"the":[25,50,87,113,116,137],"absence":[26],"a":[28,64],"tractable":[29],"trajectory":[30],"likelihood":[31,55],"hybrid":[33],"states":[35],"comprising":[36],"categorical":[37],"features":[38],"and":[39,129,144],"continuous":[40],"coordinates.":[41],"Existing":[42],"policy-optimization":[43],"strategies":[44],"address":[45],"this":[46],"difficulty":[47],"stochasticizing":[49],"generative":[51],"process":[52],"or":[53],"constructing":[54],"surrogates,":[56],"introducing":[57],"additional":[58],"approximations.":[59],"Here,":[60],"we":[61],"present":[62],"ReCoFlow,":[63,79],"novel":[65],"trajectorylikelihood-free":[66],"framework":[69],"reward-contrastive":[71],"post-training":[72],"pretrained":[74],"models.":[77],"Within":[78],"reward-favored":[80],"molecules":[81,92],"provide":[82],"positive":[83],"endpoint":[84],"supervision":[85],"model,":[89],"whereas":[90],"reward-disfavored":[91],"are":[93],"used":[94],"to":[95,104],"construct":[96],"reference-guided":[97],"negative":[98],"targets.":[99],"Additionally,":[100],"an":[101,105,160],"anchor":[102],"loss":[103],"exponential":[106],"moving":[107],"average":[108],"(EMA)":[109],"reference":[110],"model":[111],"constrains":[112],"drift":[114],"fine-tuned":[117],"model.":[118],"In":[119],"benchmarking":[120],"studies":[121],"including":[122],"physicochemical":[123],"property":[124],"optimization,":[125,128],"shape":[127],"pocket-conditioned":[130],"generation,":[133],"ReCoFlow":[134,157],"consistently":[135],"improved":[136],"performance":[138],"over":[142],"reward-weighted":[143],"surrogate-ratio":[145],"optimization":[146],"schemes":[147],"while":[148],"maintaining":[149],"high":[150],"validity.":[152],"These":[153],"results":[154],"suggest":[155],"that":[156],"could":[158],"be":[159],"effective":[161],"reinforcement-learning":[162],"strategy":[163],"goal-directed":[165],"design.":[167]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-09-10T00:00:00"}