{"id":"https://openalex.org/W4414979690","doi":"https://doi.org/10.48550/arxiv.2510.05871","title":"Towards Label-Free Biological Reasoning Synthetic Dataset Creation via Uncertainty Filtering","display_name":"Towards Label-Free Biological Reasoning Synthetic Dataset Creation via Uncertainty Filtering","publication_year":2025,"publication_date":"2025-10-07","ids":{"openalex":"https://openalex.org/W4414979690","doi":"https://doi.org/10.48550/arxiv.2510.05871"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2510.05871","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.05871","pdf_url":"https://arxiv.org/pdf/2510.05871","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/2510.05871","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119204116","display_name":"Josefa Lia Stoisser","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I189090001","display_name":"Novo Nordisk (Denmark)","ror":"https://ror.org/0435rc536","country_code":"DK","type":"company","lineage":["https://openalex.org/I189090001","https://openalex.org/I2801134892","https://openalex.org/I4210159963"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Stoisser, Josefa Lia","raw_affiliation_strings":["Novo Nordisk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Novo Nordisk","institution_ids":["https://openalex.org/I189090001"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050812342","display_name":"Lawrence Phillips","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I189090001","display_name":"Novo Nordisk (Denmark)","ror":"https://ror.org/0435rc536","country_code":"DK","type":"company","lineage":["https://openalex.org/I189090001","https://openalex.org/I2801134892","https://openalex.org/I4210159963"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Phillips, Lawrence","raw_affiliation_strings":["Novo Nordisk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Novo Nordisk","institution_ids":["https://openalex.org/I189090001"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101554186","display_name":"Aditya Misra","orcid":"https://orcid.org/0000-0003-2389-1275","observed_orcids":["https://orcid.org/0000-0003-2389-1275"]},"institutions":[{"id":"https://openalex.org/I189090001","display_name":"Novo Nordisk (Denmark)","ror":"https://ror.org/0435rc536","country_code":"DK","type":"company","lineage":["https://openalex.org/I189090001","https://openalex.org/I2801134892","https://openalex.org/I4210159963"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Misra, Aditya","raw_affiliation_strings":["Novo Nordisk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Novo Nordisk","institution_ids":["https://openalex.org/I189090001"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112933776","display_name":"Tom A. Lamb","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Lamb, Tom A.","raw_affiliation_strings":["University of Oxford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109615038","display_name":"Philip H.S. Torr","orcid":"https://orcid.org/0000-0001-5545-7217","observed_orcids":["https://orcid.org/0000-0001-5545-7217"]},"institutions":[{"id":"https://openalex.org/I40120149","display_name":"University of Oxford","ror":"https://ror.org/052gg0110","country_code":"GB","type":"education","lineage":["https://openalex.org/I40120149"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Torr, Philip","raw_affiliation_strings":["University of Oxford"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Oxford","institution_ids":["https://openalex.org/I40120149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052544681","display_name":"Marc Boubnovski Martell","orcid":"https://orcid.org/0000-0001-5379-3727","observed_orcids":["https://orcid.org/0000-0001-5379-3727"]},"institutions":[{"id":"https://openalex.org/I189090001","display_name":"Novo Nordisk (Denmark)","ror":"https://ror.org/0435rc536","country_code":"DK","type":"company","lineage":["https://openalex.org/I189090001","https://openalex.org/I2801134892","https://openalex.org/I4210159963"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Martell, Marc Boubnovski","raw_affiliation_strings":["Novo Nordisk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Novo Nordisk","institution_ids":["https://openalex.org/I189090001"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082220190","display_name":"Julien Fauqueur","orcid":"https://orcid.org/0000-0003-0227-2221","observed_orcids":["https://orcid.org/0000-0003-0227-2221"]},"institutions":[{"id":"https://openalex.org/I189090001","display_name":"Novo Nordisk (Denmark)","ror":"https://ror.org/0435rc536","country_code":"DK","type":"company","lineage":["https://openalex.org/I189090001","https://openalex.org/I2801134892","https://openalex.org/I4210159963"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"Fauqueur, Julien","raw_affiliation_strings":["Novo Nordisk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Novo Nordisk","institution_ids":["https://openalex.org/I189090001"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5062738925","display_name":"Kaspar M\u00e4rtens","orcid":"https://orcid.org/0000-0002-7631-727X","observed_orcids":["https://orcid.org/0000-0002-7631-727X"]},"institutions":[{"id":"https://openalex.org/I189090001","display_name":"Novo Nordisk (Denmark)","ror":"https://ror.org/0435rc536","country_code":"DK","type":"company","lineage":["https://openalex.org/I189090001","https://openalex.org/I2801134892","https://openalex.org/I4210159963"]}],"countries":["DK"],"is_corresponding":false,"raw_author_name":"M\u00e4rtens, Kaspar","raw_affiliation_strings":["Novo Nordisk"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Novo Nordisk","institution_ids":["https://openalex.org/I189090001"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":3,"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/T10028","display_name":"Topic Modeling","score":0.2768999934196472,"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/T10028","display_name":"Topic Modeling","score":0.2768999934196472,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.09279999881982803,"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"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.08190000057220459,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/label-free-training","display_name":"label-free training","score":0.593999981880188},{"id":"https://openalex.org/keywords/biological-reasoning","display_name":"biological reasoning","score":0.6159999966621399},{"id":"https://openalex.org/keywords/self-consistency","display_name":"self-consistency","score":0.9089999794960022},{"id":"https://openalex.org/keywords/large-reasoning-models","display_name":"large reasoning models","score":0.7120000123977661},{"id":"https://openalex.org/keywords/supervised-fine-tuning","display_name":"supervised fine-tuning","score":0.7229999899864197}],"concepts":[{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6906999945640564},{"id":"https://openalex.org/C2780505938","wikidata":"https://www.wikidata.org/wiki/Q17093282","display_name":"Unavailability","level":2,"score":0.6478000283241272},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.6087999939918518},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6008999943733215},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5777000188827515},{"id":"https://openalex.org/C100279451","wikidata":"https://www.wikidata.org/wiki/Q372193","display_name":"Perplexity","level":3,"score":0.5462999939918518},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.5026000142097473},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.49079999327659607},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.46630001068115234},{"id":"https://openalex.org/C32230216","wikidata":"https://www.wikidata.org/wiki/Q7882499","display_name":"Uncertainty quantification","level":2,"score":0.4300999939441681},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.374099999666214},{"id":"https://openalex.org/C2781170535","wikidata":"https://www.wikidata.org/wiki/Q30587856","display_name":"Noisy data","level":2,"score":0.33230000734329224},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C3020493868","wikidata":"https://www.wikidata.org/wiki/Q55631277","display_name":"Real world data","level":2,"score":0.29589998722076416},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.29339998960494995},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.28049999475479126},{"id":"https://openalex.org/C177606310","wikidata":"https://www.wikidata.org/wiki/Q5674297","display_name":"Adaptability","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2703999876976013},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.26989999413490295},{"id":"https://openalex.org/C177803969","wikidata":"https://www.wikidata.org/wiki/Q29205","display_name":"Uncertainty analysis","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.25529998540878296},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2510.05871","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.05871","pdf_url":"https://arxiv.org/pdf/2510.05871","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.2510.05871","doi":"https://doi.org/10.48550/arxiv.2510.05871","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.05871","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:2510.05871","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.05871","pdf_url":"https://arxiv.org/pdf/2510.05871","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/W4414979690.pdf","grobid_xml":"https://content.openalex.org/works/W4414979690.grobid-xml"},"referenced_works_count":19,"referenced_works":["https://openalex.org/W2965373594","https://openalex.org/W3153712677","https://openalex.org/W4221161695","https://openalex.org/W4304194220","https://openalex.org/W4385571944","https://openalex.org/W4385572634","https://openalex.org/W4398796786","https://openalex.org/W4401042444","https://openalex.org/W4402671817","https://openalex.org/W4407310145","https://openalex.org/W4414281281","https://openalex.org/W4414922014","https://openalex.org/W4415338080","https://openalex.org/W4416131312","https://openalex.org/W4416260146","https://openalex.org/W4416711605","https://openalex.org/W6948393355","https://openalex.org/W6967242028","https://openalex.org/W7133192525"],"related_works":["https://openalex.org/W7154608054","https://openalex.org/W7117717802","https://openalex.org/W4414922014","https://openalex.org/W7133350380","https://openalex.org/W7133341380","https://openalex.org/W4415338080","https://openalex.org/W4402712382","https://openalex.org/W4416445771","https://openalex.org/W7127288381","https://openalex.org/W4415199731"],"abstract_inverted_index":{"Synthetic":[0],"chain-of-thought":[1],"(CoT)":[2],"traces":[3,30,79],"are":[4,42,95],"widely":[5],"used":[6],"to":[7,25,86,122],"train":[8],"large":[9],"reasoning":[10,78,160],"models":[11],"(LRMs),":[12],"improving":[13],"generalization":[14],"by":[15],"providing":[16],"step-level":[17],"supervision.":[18],"Yet":[19],"most":[20],"approaches":[21],"require":[22],"ground-truth":[23,123],"labels":[24,94],"seed":[26],"or":[27],"filter":[28],"these":[29],"-":[31,57,68],"an":[32],"expensive":[33],"bottleneck":[34],"in":[35,165],"domains":[36,166],"like":[37,63],"biology":[38],"where":[39,92,167],"wet-lab":[40,93],"data":[41,114],"scarce.":[43],"We":[44,75],"propose":[45],"a":[46,53,70,90,155],"label-free":[47],"alternative:":[48],"uncertainty-based":[49],"filtering,":[50],"which":[51],"uses":[52],"model's":[54],"own":[55],"confidence":[56,153],"quantified":[58],"through":[59],"established":[60],"uncertainty":[61,138,143],"metrics":[62,144],"self-consistency":[64],"and":[65,80,107,125,140],"predictive":[66],"perplexity":[67],"as":[69],"substitute":[71],"for":[72,136,158],"external":[73],"labels.":[74],"sample":[76],"multiple":[77],"retain":[81],"only":[82],"low-uncertainty":[83],"subsets.":[84],"Applied":[85],"biological":[87],"perturbation":[88],"prediction,":[89],"domain":[91],"especially":[96],"costly,":[97],"we":[98],"show":[99,131],"that":[100,108,132,141,151],"the":[101,120],"filtered":[102],"subset":[103],"has":[104],"higher":[105],"accuracy,":[106],"supervised":[109],"fine-tuning":[110],"(SFT)":[111],"on":[112],"uncertainty-filtered":[113],"outperforms":[115],"unfiltered":[116],"synthetic":[117],"data,":[118],"narrows":[119],"gap":[121],"training,":[124],"surpasses":[126],"strong":[127],"LRM":[128],"baselines.":[129],"Ablations":[130],"per-class":[133],"filtering":[134],"corrects":[135],"class-specific":[137],"scales":[139],"hybrid":[142],"yield":[145],"higher-quality":[146],"datasets.":[147],"Our":[148],"results":[149],"suggest":[150],"model-internal":[152],"is":[154,169],"powerful":[156],"signal":[157],"efficient":[159],"dataset":[161],"creation,":[162],"enabling":[163],"LRMs":[164],"supervision":[168],"expensive.":[170]},"counts_by_year":[{"year":2026,"cited_by_count":3}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}