{"id":"https://openalex.org/W7204904671","doi":"https://doi.org/10.26434/chemrxiv.15008164/v1","title":"Transfer Protocol Matters More Than Source-Task Selection in Low-Data Molecular Property Prediction: An Audit of QM9 Pretraining on Computed Oxidation Potentials and Redox Reaction Energies","display_name":"Transfer Protocol Matters More Than Source-Task Selection in Low-Data Molecular Property Prediction: An Audit of QM9 Pretraining on Computed Oxidation Potentials and Redox Reaction Energies","publication_year":2026,"publication_date":"2026-09-01","ids":{"openalex":"https://openalex.org/W7204904671","doi":"https://doi.org/10.26434/chemrxiv.15008164/v1"},"language":"en","primary_location":{"id":"doi:10.26434/chemrxiv.15008164/v1","doi":"https://doi.org/10.26434/chemrxiv.15008164/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15008164/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008164/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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","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.15008164/v1","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138874373","display_name":"Yueru Zhang","orcid":"https://orcid.org/0009-0000-3912-3882","observed_orcids":["https://orcid.org/0009-0000-3912-3882"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yueru Zhang","raw_affiliation_strings":[],"raw_orcid":"https://orcid.org/0009-0000-3912-3882","affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5138874373"],"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/T11948","display_name":"Machine Learning in Materials Science","score":0.5913000106811523,"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":0.5913000106811523,"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":0.32249999046325684,"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/T10002","display_name":"Advanced Chemical Physics Studies","score":0.06629999727010727,"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/oxidation-potential","display_name":"oxidation potential","score":0.6000000238418579},{"id":"https://openalex.org/keywords/task-similarity","display_name":"task similarity","score":0.531000018119812},{"id":"https://openalex.org/keywords/feature-overlap","display_name":"feature overlap","score":0.34200000762939453},{"id":"https://openalex.org/keywords/magnitude-pruning","display_name":"magnitude pruning","score":0.5339999794960022},{"id":"https://openalex.org/keywords/transfer-protocols","display_name":"transfer protocols","score":0.39100000262260437},{"id":"https://openalex.org/keywords/negative-transfer","display_name":"negative transfer","score":0.43799999356269836},{"id":"https://openalex.org/keywords/pseudo-labeling","display_name":"pseudo-labeling","score":0.492000013589859},{"id":"https://openalex.org/keywords/interpretability","display_name":"interpretability","score":0.5040000081062317}],"concepts":[{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.6229000091552734},{"id":"https://openalex.org/C2776175482","wikidata":"https://www.wikidata.org/wiki/Q1195816","display_name":"Transfer (computing)","level":2,"score":0.5577999949455261},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5462999939918518},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5360000133514404},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.48820000886917114},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.47769999504089355},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.4645000100135803},{"id":"https://openalex.org/C55904794","wikidata":"https://www.wikidata.org/wiki/Q82682","display_name":"Redox","level":2,"score":0.44290000200271606},{"id":"https://openalex.org/C2991733762","wikidata":"https://www.wikidata.org/wiki/Q82682","display_name":"Oxidation reduction","level":2,"score":0.4422999918460846},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4345000088214874},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.43160000443458557},{"id":"https://openalex.org/C186060115","wikidata":"https://www.wikidata.org/wiki/Q30336093","display_name":"Biological system","level":1,"score":0.4142000079154968},{"id":"https://openalex.org/C10390740","wikidata":"https://www.wikidata.org/wiki/Q2282401","display_name":"Analyte","level":2,"score":0.3937000036239624},{"id":"https://openalex.org/C189950617","wikidata":"https://www.wikidata.org/wiki/Q937228","display_name":"Property (philosophy)","level":2,"score":0.3587999939918518},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3449999988079071},{"id":"https://openalex.org/C148483581","wikidata":"https://www.wikidata.org/wiki/Q446488","display_name":"Feature selection","level":2,"score":0.3425000011920929},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.32739999890327454},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3206999897956848},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.31700000166893005},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.30559998750686646},{"id":"https://openalex.org/C106447425","wikidata":"https://www.wikidata.org/wiki/Q11379","display_name":"Energy transfer","level":2,"score":0.30489999055862427},{"id":"https://openalex.org/C143724316","wikidata":"https://www.wikidata.org/wiki/Q312468","display_name":"Series (stratigraphy)","level":2,"score":0.2996000051498413},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2919999957084656},{"id":"https://openalex.org/C2779227376","wikidata":"https://www.wikidata.org/wiki/Q6505497","display_name":"Layer (electronics)","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.25369998812675476},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.25119999051094055},{"id":"https://openalex.org/C2777363581","wikidata":"https://www.wikidata.org/wiki/Q15098235","display_name":"Harm","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.26434/chemrxiv.15008164/v1","doi":"https://doi.org/10.26434/chemrxiv.15008164/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15008164/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008164/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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":null,"raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.26434/chemrxiv.15008164/v1","doi":"https://doi.org/10.26434/chemrxiv.15008164/v1","is_oa":true,"landing_page_url":"https://doi.org/10.26434/chemrxiv.15008164/v1","pdf_url":"https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15008164/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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","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":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204904671.pdf","grobid_xml":"https://content.openalex.org/works/W7204904671.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W1845051632","https://openalex.org/W2080635178","https://openalex.org/W2149933564","https://openalex.org/W2165698076","https://openalex.org/W2169393322","https://openalex.org/W2212983101","https://openalex.org/W2594633041","https://openalex.org/W2805003733","https://openalex.org/W2942810103","https://openalex.org/W2980113592","https://openalex.org/W4310195198","https://openalex.org/W4410822772"],"related_works":["https://openalex.org/W7165152734","https://openalex.org/W7162301920","https://openalex.org/W4310820019","https://openalex.org/W3119174773","https://openalex.org/W4391539102","https://openalex.org/W4394890631","https://openalex.org/W6987585224","https://openalex.org/W7167740683","https://openalex.org/W7071683529","https://openalex.org/W3032781902"],"abstract_inverted_index":{"Pretraining":[0],"a":[1,11,16,24,78,115,145],"network":[2],"on":[3,42,158],"QM9":[4,67],"quantum-chemical":[5,72],"labels":[6],"and":[7,23,55,112,140,183,230,249,279],"transferring":[8],"it":[9],"to":[10,35,128,233],"data-scarce":[12],"molecular":[13,161],"endpoint":[14],"is":[15],"common":[17],"strategy":[18],"in":[19,77,134,144,192,253],"chemical":[20],"machine":[21],"learning,":[22],"growing":[25],"literature":[26],"proposes":[27],"tasksimilarity":[28],"descriptors":[29],"for":[30,108,114],"deciding":[31],"which":[32],"source":[33,68,156],"task":[34],"transfer":[36,168,172,205,208,232,268,277],"from.":[37],"We":[38,74,196],"audit":[39],"both":[40],"decisions":[41],"two":[43,160],"non-synthetic":[44],"computational":[45],"chemistry":[46],"datasets":[47,162],"\u2014":[48,62],"DFT-derived":[49],"oxidation":[50],"potentials":[51],"(OxPot,":[52,269],"15,238":[53],"molecules)":[54,61],"redox":[56],"reaction":[57],"energies":[58],"(RedDB,":[59],"15,551":[60],"using":[63],"24":[64],"independently":[65],"initialised":[66],"models":[69],"spanning":[70],"eight":[71],"tasks.":[73],"first":[75,235],"establish,":[76],"controlled":[79],"synthetic":[80],"benchmark":[81],"with":[82,130],"known":[83],"informative":[84],"coordinates,":[85],"that":[86,97],"magnitude-pruning":[87],"masks":[88],"can":[89],"be":[90],"aggregated":[91],"into":[92,154],"interpretable":[93],"task-level":[94],"feature-usage":[95],"signatures":[96],"recover":[98],"ground-truth":[99],"feature":[100],"overlap":[101],"(pooled":[102],"mean":[103,167,267],"Spearman":[104],"0.2749,":[105],"versus":[106],"0.1755":[107],"unpruned":[109],"first-layer":[110],"weights":[111],"0.0614":[113],"random":[116],"control).":[117],"Eleven":[118],"signature":[119],"series":[120],"were":[121],"compared;":[122],"their":[123],"relative":[124,219],"ordering":[125],"proved":[126],"sensitive":[127],"pooling,":[129],"gradient":[131],"saliency":[132],"ahead":[133],"the":[135,159,211,215,234,247,258],"unmatched":[136],"pooled":[137],"summary":[138],"(0.2855)":[139],"behind":[141],"by":[142,190,221],"0.211":[143],"paired":[146],"same-setting":[147],"audit.":[148],"This":[149],"mechanism":[150],"did":[151,262],"not":[152],"translate":[153],"deployable":[155],"selection:":[157],"every":[163],"selector":[164],"produced":[165,246],"negative":[166],"gain":[169,220],"under":[170,203,257],"full-weight":[171],"(feature-similarity":[173],"top-1:":[174,180],"\u22120.124,":[175],"negative-transfer":[176],"rate":[177],"95.8%;":[178],"pseudo-label":[179],"\u22120.177,":[181],"100%),":[182],"an":[184,280],"agreement":[185],"gate":[186],"limited":[187],"harm":[188],"mainly":[189],"abstaining":[191],"87.5%":[193],"of":[194],"cases.":[195],"then":[197],"evaluated":[198],"144":[199],"matched":[200],"case\u2013source":[201],"pairs":[202],"four":[204],"protocols":[206],"(576":[207],"runs).":[209],"At":[210],"target-case":[212],"level,":[213],"resetting":[214],"output":[216],"head":[217],"improved":[218],"0.087":[222],"(case-bootstrap":[223],"95%":[224],"CI":[225,240],"0.070\u20130.106;":[226],"18/18":[227,242],"cases":[228],"positive),":[229],"restricting":[231],"layer":[236],"added":[237],"0.044":[238],"(95%":[239],"0.034\u20130.055;":[241],"positive).":[243],"Protocol":[244],"changes":[245],"largest":[248],"most":[250,259],"consistent":[251],"shifts":[252],"realised":[254],"gain.":[255],"Only":[256],"conservative":[260],"protocol":[261,278],"any":[263],"dataset":[264],"reach":[265],"non-negative":[266],"+0.003;":[270],"RedDB":[271],"remained":[272],"negative).":[273],"For":[274],"this":[275],"setting,":[276],"explicit":[281],"abstention":[282],"option":[283],"deserve":[284],"priority":[285],"over":[286],"increasingly":[287],"elaborate":[288],"sourcesimilarity":[289],"descriptors.":[290]},"counts_by_year":[],"updated_date":"2026-11-04T10:54:05.112534","created_date":"2026-09-02T00:00:00"}