{"id":"https://openalex.org/W7160175026","doi":"https://doi.org/10.48550/arxiv.2605.00640","title":"Knowing when to trust machine-learned interatomic potentials","display_name":"Knowing when to trust machine-learned interatomic potentials","publication_year":2026,"publication_date":"2026-05-01","ids":{"openalex":"https://openalex.org/W7160175026","doi":"https://doi.org/10.48550/arxiv.2605.00640"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.00640","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.00640","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.00640","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5073978740","display_name":"Shams Mehdi","orcid":"https://orcid.org/0000-0002-4078-7501"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mehdi, Shams","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090850960","display_name":"Ilkwon Cho","orcid":"https://orcid.org/0000-0002-7279-7384"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cho, Ilkwon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135100221","display_name":"Olexandr Isayev","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Isayev, Olexandr","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":0.9939000010490417,"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":0.9939000010490417,"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":0.0013000000035390258,"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/T10030","display_name":"Electrocatalysts for Energy Conversion","score":0.0005000000237487257,"subfield":{"id":"https://openalex.org/subfields/2105","display_name":"Renewable Energy, Sustainability and the Environment"},"field":{"id":"https://openalex.org/fields/21","display_name":"Energy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.6830999851226807},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.6801000237464905},{"id":"https://openalex.org/keywords/discriminative-model","display_name":"Discriminative model","score":0.6237000226974487},{"id":"https://openalex.org/keywords/uncertainty-quantification","display_name":"Uncertainty quantification","score":0.5963000059127808},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.5703999996185303},{"id":"https://openalex.org/keywords/monotonic-function","display_name":"Monotonic function","score":0.525600016117096},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.49070000648498535},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.43939998745918274},{"id":"https://openalex.org/keywords/measurement-uncertainty","display_name":"Measurement uncertainty","score":0.3264000117778778}],"concepts":[{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.6830999851226807},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.6801000237464905},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.6237000226974487},{"id":"https://openalex.org/C32230216","wikidata":"https://www.wikidata.org/wiki/Q7882499","display_name":"Uncertainty quantification","level":2,"score":0.5963000059127808},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.5703999996185303},{"id":"https://openalex.org/C72169020","wikidata":"https://www.wikidata.org/wiki/Q194404","display_name":"Monotonic function","level":2,"score":0.525600016117096},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5242999792098999},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.49070000648498535},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.4733000099658966},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.45649999380111694},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.43939998745918274},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.41359999775886536},{"id":"https://openalex.org/C137209882","wikidata":"https://www.wikidata.org/wiki/Q1403517","display_name":"Measurement uncertainty","level":2,"score":0.3264000117778778},{"id":"https://openalex.org/C170797541","wikidata":"https://www.wikidata.org/wiki/Q452399","display_name":"Error bar","level":2,"score":0.3208000063896179},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.31619998812675476},{"id":"https://openalex.org/C66905080","wikidata":"https://www.wikidata.org/wiki/Q17005494","display_name":"Binary classification","level":3,"score":0.3010999858379364},{"id":"https://openalex.org/C100253034","wikidata":"https://www.wikidata.org/wiki/Q196372","display_name":"Systematic error","level":2,"score":0.29440000653266907},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.29100000858306885},{"id":"https://openalex.org/C59593255","wikidata":"https://www.wikidata.org/wiki/Q901663","display_name":"Molecular dynamics","level":2,"score":0.2872999906539917},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.28519999980926514},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2806999981403351},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.28040000796318054},{"id":"https://openalex.org/C19619285","wikidata":"https://www.wikidata.org/wiki/Q196372","display_name":"Observational error","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C186060115","wikidata":"https://www.wikidata.org/wiki/Q30336093","display_name":"Biological system","level":1,"score":0.26350000500679016},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2621999979019165},{"id":"https://openalex.org/C2779190172","wikidata":"https://www.wikidata.org/wiki/Q4913888","display_name":"Binary data","level":3,"score":0.2612000107765198},{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.25839999318122864}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.00640","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.00640","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.00640","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.00640","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":"Preprint"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.7534940242767334}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Prevailing":[0],"machine-learned":[1],"interatomic":[2],"potential":[3],"(MLIP)":[4],"uncertainty-quantification":[5],"methods":[6,15],"rely":[7],"on":[8,145],"ensembles":[9],"of":[10,38,108],"independently":[11],"trained":[12],"backbones.":[13],"These":[14],"scale":[16],"unfavorably":[17],"with":[18,27,42,105],"foundation-scale":[19,118],"MLIPs,":[20],"and":[21,88,139,141],"their":[22],"member-disagreement":[23],"signals":[24],"correlate":[25],"weakly":[26],"per-molecule":[28],"prediction":[29],"error.":[30],"Here":[31],"we":[32],"probe":[33],"the":[34,80,106,109],"frozen":[35],"per-atom":[36,124,150],"representations":[37],"a":[39,43,68,99,113],"pretrained":[40],"MLIP":[41,48,92,147],"compact":[44],"discriminative":[45],"classifier,":[46],"recasting":[47],"uncertainty":[49],"quantification":[50],"as":[51,98],"selective":[52],"classification":[53],"rather":[54],"than":[55],"error":[56,76],"regression.":[57],"The":[58],"resulting":[59],"method,":[60],"PROBE":[61,94,136],"(Post-hoc":[62],"Reliability":[63],"frOm":[64],"Backbone":[65],"Embeddings),":[66],"produces":[67],"per-prediction":[69],"reliability":[70,101],"probability":[71],"that":[72,148],"monotonically":[73],"tracks":[74],"actual":[75],"without":[77],"modification":[78],"to":[79],"underlying":[81],"model.":[82],"Across":[83],"large":[84],"held-out":[85],"evaluation":[86],"sets":[87],"two":[89],"structurally":[90],"distinct":[91],"architectures,":[93],"outperforms":[95],"ensemble":[96],"disagreement":[97],"binary":[100],"signal,":[102],"which":[103],"strengthens":[104],"expressiveness":[107],"backbone":[110],"representation,":[111],"implying":[112],"favorable":[114],"scaling":[115],"trajectory":[116],"toward":[117],"MLIPs.":[119],"Multi-head":[120],"self-attention":[121],"additionally":[122],"yields":[123],"importance":[125],"maps,":[126],"providing":[127],"chemically":[128],"interpretable":[129],"diagnostics":[130],"at":[131],"no":[132],"additional":[133],"computational":[134],"cost.":[135],"is":[137,142],"post-hoc":[138],"architecture-agnostic,":[140],"directly":[143],"deployable":[144],"any":[146],"exposes":[149],"representations.":[151]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-05T00:00:00"}