{"id":"https://openalex.org/W7214664710","doi":"https://doi.org/10.26434/chemrxiv.15009537/v1","title":"Auditing LLM Chemical Priors for Multi-Objective Bayesian Optimization of Chemical Reactions: A Validity, Alignment, and Diversity Benchmark","display_name":"Auditing LLM Chemical Priors for Multi-Objective Bayesian Optimization of Chemical Reactions: A Validity, Alignment, and Diversity Benchmark","publication_year":2026,"publication_date":"2026-09-28","ids":{"openalex":"https://openalex.org/W7214664710","doi":"https://doi.org/10.26434/chemrxiv.15009537/v1"},"language":"en","primary_location":{"id":"doi:10.26434/chemrxiv.15009537/v1","doi":"https://doi.org/10.26434/chemrxiv.15009537/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15009537/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15009537/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.15009537/v1","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153190942","display_name":"Fan Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Fan Zhang","raw_affiliation_strings":["Independent Researcher"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent Researcher","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5153190942"],"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/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.801800012588501,"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"}},"topics":[{"id":"https://openalex.org/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.801800012588501,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.14810000360012054,"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/T11798","display_name":"Optimal Experimental Design Methods","score":0.017000000923871994,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8880000114440918},{"id":"https://openalex.org/keywords/bayesian-optimization","display_name":"Bayesian optimization","score":0.781000018119812},{"id":"https://openalex.org/keywords/multi-objective-optimization","display_name":"multi-objective optimization","score":0.9319999814033508},{"id":"https://openalex.org/keywords/chemical-reaction-optimization","display_name":"chemical reaction optimization","score":0.9340000152587891},{"id":"https://openalex.org/keywords/suzuki-miyaura-coupling","display_name":"Suzuki coupling","score":0.8119999766349792},{"id":"https://openalex.org/keywords/buchwald-hartwig-amination","display_name":"Buchwald-Hartwig amination","score":0.9399999976158142},{"id":"https://openalex.org/keywords/hypervolume","display_name":"hypervolume","score":0.546999990940094},{"id":"https://openalex.org/keywords/sampling-efficiency","display_name":"sampling efficiency","score":0.7910000085830688},{"id":"https://openalex.org/keywords/farthest-point-sampling","display_name":"farthest point sampling","score":0.8209999799728394},{"id":"https://openalex.org/keywords/diversity-control","display_name":"diversity control","score":0.5600000023841858}],"concepts":[{"id":"https://openalex.org/C114466953","wikidata":"https://www.wikidata.org/wiki/Q6034165","display_name":"Initialization","level":2,"score":0.8533999919891357},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.8263999819755554},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6251000165939331},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.61080002784729},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.48489999771118164},{"id":"https://openalex.org/C173801870","wikidata":"https://www.wikidata.org/wiki/Q201413","display_name":"Heuristic","level":2,"score":0.3765999972820282},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3587999939918518},{"id":"https://openalex.org/C207201462","wikidata":"https://www.wikidata.org/wiki/Q182505","display_name":"Bayes' theorem","level":3,"score":0.326200008392334},{"id":"https://openalex.org/C111350023","wikidata":"https://www.wikidata.org/wiki/Q1191869","display_name":"Markov chain Monte Carlo","level":3,"score":0.3138999938964844},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.2962999939918518},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.296099990606308},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2944999933242798},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28679999709129333},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.27950000762939453},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.26489999890327454},{"id":"https://openalex.org/C41426520","wikidata":"https://www.wikidata.org/wiki/Q1192065","display_name":"Point estimation","level":2,"score":0.26010000705718994},{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.2515000104904175},{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.26434/chemrxiv.15009537/v1","doi":"https://doi.org/10.26434/chemrxiv.15009537/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15009537/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15009537/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.15009537/v1","doi":"https://doi.org/10.26434/chemrxiv.15009537/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15009537/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15009537/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":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214664710.pdf","grobid_xml":"https://content.openalex.org/works/W7214664710.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W4306845856","https://openalex.org/W4389991792","https://openalex.org/W4396723768","https://openalex.org/W4396852083","https://openalex.org/W4399837985","https://openalex.org/W4405185373","https://openalex.org/W4407170749","https://openalex.org/W4409656093","https://openalex.org/W4411900553","https://openalex.org/W4416036529","https://openalex.org/W4417070828","https://openalex.org/W4417182937","https://openalex.org/W4417517279","https://openalex.org/W6979288382"],"related_works":["https://openalex.org/W7151074652","https://openalex.org/W7079505566","https://openalex.org/W7163224156","https://openalex.org/W4415836214","https://openalex.org/W6979288382","https://openalex.org/W7220444821","https://openalex.org/W7203826422","https://openalex.org/W4417070828","https://openalex.org/W7214022771","https://openalex.org/W7083693793"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"are":[4,131,218],"increasingly":[5],"proposed":[6],"as":[7],"priors":[8,130,217],"for":[9,224,235],"cold-starting":[10],"Bayesian":[11],"optimization":[12,24],"(BO)":[13],"of":[14],"chemical":[15],"reactions,":[16],"yet":[17,220],"their":[18],"benefit\u2013risk":[19],"profile":[20],"in":[21],"multi-objective":[22,42,225],"reaction":[23,48,226],"remains":[25],"poorly":[26],"characterized.":[27],"We":[28],"present":[29],"a":[30,97,137,152,162,221,230],"systematic":[31],"benchmark":[32],"asking":[33],"when":[34,237],"LLM":[35,106,129,140,183,216],"knowledge":[36],"helps,":[37],"is":[38],"neutral,":[39],"or":[40,181],"hurts":[41],"BO":[43,99],"(MOBO).":[44],"Across":[45],"seven":[46],"public":[47],"benchmarks":[49],"(6,384":[50],"candidate":[51],"experiments),":[52],"we":[53,228],"compare":[54],"nine":[55],"strategies\u2014including":[56],"LLM-initialized":[57],"BO,":[58],"an":[59],"LLM-only":[60],"initializer,":[61],"and":[62,83,92,188],"established":[63],"baselines":[64],"(TSEMO,":[65],"EDBO+-style":[66],"scalarization,":[67],"LLAMBO)\u2014with":[68],"20":[69],"seeds":[70],"per":[71],"configuration":[72],"(1,120":[73],"main-grid":[74],"plus":[75],"360":[76],"ablation":[77],"campaigns),":[78],"scored":[79],"by":[80],"hypervolume":[81],"completion":[82],"its":[84,169],"sampling-efficiency":[85],"integral":[86],"(AUC),":[87],"with":[88,161],"paired":[89],"Wilcoxon":[90],"tests":[91],"two-family":[93],"Benjamini\u2013Hochberg":[94],"correction":[95],"against":[96],"random-init":[98],"control.":[100],"Three":[101],"regimes":[102],"emerge.":[103],"On":[104,127,136],"Suzuki\u2013Miyaura,":[105],"initialization":[107,141,176,184],"significantly":[108,142],"improves":[109],"sampling":[110],"efficiency":[111,144],"(\u0394AUC":[112,145],"+2.0":[113],"points,":[114,121,147],"q":[115,122,148],"=":[116,123,149],"0.026;":[117],"alternative":[118],"backbone":[119],"+2.7":[120],"1.4":[124],"\u00d7":[125],"10\u207b\u2074).":[126],"Buchwald\u2013Hartwig,":[128],"statistically":[132],"indistinguishable":[133],"from":[134],"random.":[135],"process-development":[138],"coupling,":[139],"harms":[143],"\u221211.1":[146],"0.0016)":[150],"despite":[151],"0%":[153],"out-of-pool":[154,164,206],"rate,":[155],"whereas":[156],"the":[157,173,194,209],"Suzuki":[158],"benefit":[159,191],"coexists":[160],"45%":[163],"rate\u2014the":[165],"prior\u2019s":[166],"direction,":[167],"not":[168,219],"syntactic":[170],"validity,":[171],"controls":[172],"outcome.":[174],"Diversity-based":[175],"(farthest":[177],"point":[178],"sampling)":[179],"matches":[180],"exceeds":[182,192],"wherever":[185],"LLMs":[186],"help,":[187],"no":[189],"LLM-derived":[190],"what":[193],"knowledge-free":[195],"diversity":[196],"control":[197],"already":[198],"provides.":[199],"A":[200],"five-category":[201],"diagnostic":[202],"protocol":[203],"intercepts":[204],"100%":[205],"suggestions":[207],"on":[208],"flow":[210],"cases":[211],"at":[212],"zero":[213],"experimental":[214],"cost.":[215],"reliable":[222],"default":[223],"optimization;":[227],"provide":[229],"cost-free":[231],"pre-campaign":[232],"screening":[233],"heuristic":[234],"deciding":[236],"to":[238],"trust":[239],"them.":[240]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-09-29T00:00:00"}