{"id":"https://openalex.org/W7204455734","doi":"https://doi.org/10.48550/arxiv.2608.26086","title":"TraceML: What Auto-Research Agents Miss in Long-Horizon ML Development","display_name":"TraceML: What Auto-Research Agents Miss in Long-Horizon ML Development","publication_year":2026,"publication_date":"2026-08-26","ids":{"openalex":"https://openalex.org/W7204455734","doi":"https://doi.org/10.48550/arxiv.2608.26086"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.26086","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26086","pdf_url":"https://arxiv.org/pdf/2608.26086","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.26086","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148463959","display_name":"Jiarui Yan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yan, Jiarui","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5126241244","display_name":"Weiwei Sun","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sun, Weiwei","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148429033","display_name":"Sijie Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li, Sijie","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148435110","display_name":"Wenhan Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Li, Wenhan","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5126869042","display_name":"Yiming Yang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yang, Yiming","raw_affiliation_strings":["Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I74973139"],"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/T12535","display_name":"Machine Learning and Data Classification","score":0.5562000274658203,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T12535","display_name":"Machine Learning and Data Classification","score":0.5562000274658203,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"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/T10260","display_name":"Software Engineering Research","score":0.22669999301433563,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T11986","display_name":"Scientific Computing and Data Management","score":0.0348999984562397,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"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/machine-learning-competition","display_name":"machine learning competition","score":0.6610000133514404},{"id":"https://openalex.org/keywords/human-agent-comparison","display_name":"human-agent comparison","score":0.7179999947547913},{"id":"https://openalex.org/keywords/human-ai-collaboration","display_name":"human-AI collaboration","score":0.5109999775886536},{"id":"https://openalex.org/keywords/codex","display_name":"codices","score":0.6140000224113464},{"id":"https://openalex.org/keywords/kaggle","display_name":"Kaggle","score":0.5619999766349792}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7639999985694885},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6973000168800354},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.6284999847412109},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.61080002784729},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.607200026512146},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.567300021648407},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5087000131607056},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.48559999465942383},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.39160001277923584},{"id":"https://openalex.org/C2776542497","wikidata":"https://www.wikidata.org/wiki/Q5266672","display_name":"Development (topology)","level":2,"score":0.3382999897003174},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.3124000132083893},{"id":"https://openalex.org/C148524875","wikidata":"https://www.wikidata.org/wiki/Q6975395","display_name":"F1 score","level":2,"score":0.30390000343322754},{"id":"https://openalex.org/C198352243","wikidata":"https://www.wikidata.org/wiki/Q37105","display_name":"Line (geometry)","level":2,"score":0.27619999647140503},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.267300009727478}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.26086","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26086","pdf_url":"https://arxiv.org/pdf/2608.26086","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.26086","doi":"https://doi.org/10.48550/arxiv.2608.26086","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.26086","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.26086","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.26086","pdf_url":"https://arxiv.org/pdf/2608.26086","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204455734.pdf","grobid_xml":"https://content.openalex.org/works/W7204455734.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W4403345800","https://openalex.org/W4404354143","https://openalex.org/W4417159298","https://openalex.org/W7163647702","https://openalex.org/W7196932832"],"related_works":["https://openalex.org/W4407126828","https://openalex.org/W7213526916","https://openalex.org/W7139054217","https://openalex.org/W7203950721","https://openalex.org/W7160635469","https://openalex.org/W7215103926","https://openalex.org/W4416953356","https://openalex.org/W7172027503","https://openalex.org/W7169827629","https://openalex.org/W7134250124"],"abstract_inverted_index":{"Auto-research":[0],"agents":[1],"now":[2],"run":[3,232],"machine-learning":[4],"development":[5,48],"unattended":[6],"for":[7,104],"hours,":[8],"revising":[9],"data":[10,126],"pipelines,":[11],"models,":[12,223],"and":[13,45,58,90,102,113,131,133,156,165,193,220,241],"validation":[14],"from":[15,181,233],"their":[16],"own":[17],"feedback,":[18],"yet":[19],"on":[20,61],"most":[21],"competitions":[22,64],"they":[23,40,137],"still":[24],"finish":[25],"below":[26],"strong":[27],"human":[28,57,70,89,170,182,191,246],"competitors.":[29],"Outcome-based":[30],"benchmarks":[31],"record":[32],"this":[33,118],"gap":[34,121,209],"but":[35,196],"not":[36],"its":[37,98,100,108,152,161],"cause,":[38],"because":[39],"grade":[41],"the":[42,47,62,105,110,114,120,169,185,190,197,205,208,216,218,221,245],"final":[43],"submission":[44],"discard":[46],"process":[49],"behind":[50],"it.":[51],"We":[52,214],"introduce":[53],"TraceML,":[54],"which":[55,78],"pairs":[56],"agent":[59,84,92,142,236],"work":[60],"same":[63],"under":[65],"one":[66],"version-level":[67],"schema:":[68],"4{,}465":[69],"Kaggle":[71],"trajectories":[72],"across":[73],"134":[74],"competitions,":[75],"seven":[76],"of":[77,207],"are":[79],"also":[80],"worked":[81],"by":[82],"two":[83],"scaffolds,":[85],"giving":[86],"430":[87],"paired":[88],"207":[91],"trajectories.":[93],"Every":[94],"code":[95],"version":[96],"carries":[97],"score,":[99],"timestamp,":[101],"labels":[103],"action":[106],"taken,":[107],"intent,":[109],"edit":[111],"size,":[112],"score":[115],"effect.":[116],"Read":[117],"way,":[119],"becomes":[122],"concrete.":[123],"Experts":[124],"alternate":[125],"work,":[127],"validation,":[128],"model":[129,162],"changes,":[130],"ensembling,":[132],"return":[134],"to":[135,212],"approaches":[136],"had":[138],"set":[139],"aside.":[140],"Each":[141],"scaffold":[143],"instead":[144],"collapses":[145],"into":[146,237],"a":[147,231,238],"narrow":[148],"loop:":[149],"Codex":[150],"spends":[151],"steps":[153],"re-weighting":[154],"ensembles":[155],"tuning":[157],"submissions,":[158],"MLEvolve":[159],"mutates":[160],"in":[163],"place,":[164],"neither":[166],"pivots":[167],"at":[168],"rate":[171],"nor":[172],"reopens":[173],"abandoned":[174],"work.":[175],"A":[176],"short":[177],"planning":[178],"prompt":[179],"distilled":[180],"practice":[183],"moves":[184],"behaviors":[186],"it":[187,243],"names":[188],"toward":[189],"profile":[192,199],"lifts":[194],"scores,":[195],"effort":[198],"stays":[200],"agent-shaped:":[201],"instruction":[202],"closes":[203],"only":[204],"part":[206],"that":[210,229],"reduces":[211],"instructions.":[213],"release":[215],"corpus,":[217],"schema,":[219],"labeling":[222],"together":[224],"with":[225],"an":[226],"open-source":[227],"toolkit":[228],"turns":[230],"any":[234],"command-line":[235],"TraceML":[239],"trajectory":[240],"reads":[242],"against":[244],"cohorts.":[247]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-08-28T00:00:00"}