{"id":"https://openalex.org/W7160592542","doi":"https://doi.org/10.48550/arxiv.2605.06047","title":"TFM-Retouche: A Lightweight Input-Space Adapter for Tabular Foundation Models","display_name":"TFM-Retouche: A Lightweight Input-Space Adapter for Tabular Foundation Models","publication_year":2026,"publication_date":"2026-05-07","ids":{"openalex":"https://openalex.org/W7160592542","doi":"https://doi.org/10.48550/arxiv.2605.06047"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.06047","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.06047","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.06047","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135704361","display_name":"Duong Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nguyen, Duong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135721076","display_name":"Mohammed Jawhar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jawhar, Mohammed","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5060566931","display_name":"Nicolas Chesneau","orcid":"https://orcid.org/0000-0002-8313-5829"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chesneau, Nicolas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.23919999599456787,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.23919999599456787,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.164000004529953,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.10350000113248825,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/adapter","display_name":"Adapter (computing)","score":0.8725000023841858},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.5460000038146973},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5005999803543091},{"id":"https://openalex.org/keywords/guard","display_name":"Guard (computer science)","score":0.366100013256073},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.34599998593330383},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.33970001339912415}],"concepts":[{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.8725000023841858},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7265999913215637},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.5460000038146973},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.527999997138977},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5005999803543091},{"id":"https://openalex.org/C141141315","wikidata":"https://www.wikidata.org/wiki/Q2379942","display_name":"Guard (computer science)","level":2,"score":0.366100013256073},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.35830000042915344},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.34599998593330383},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.33970001339912415},{"id":"https://openalex.org/C154504017","wikidata":"https://www.wikidata.org/wiki/Q853614","display_name":"Identifier","level":2,"score":0.3280999958515167},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.32359999418258667},{"id":"https://openalex.org/C207850805","wikidata":"https://www.wikidata.org/wiki/Q269608","display_name":"Reverse engineering","level":2,"score":0.3052000105381012},{"id":"https://openalex.org/C115051666","wikidata":"https://www.wikidata.org/wiki/Q6522493","display_name":"Ranging","level":2,"score":0.30329999327659607},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2824000120162964},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.2766999900341034},{"id":"https://openalex.org/C190839683","wikidata":"https://www.wikidata.org/wiki/Q2448197","display_name":"Train","level":2,"score":0.2621000111103058},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.25859999656677246},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.25690001249313354}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.06047","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.06047","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.06047","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.06047","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":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Tabular":[0],"foundation":[1],"models":[2],"(TFMs),":[3],"such":[4,53],"as":[5,54],"TabPFN-2.6,":[6],"TabICLv2,":[7],"ConTextTab,":[8],"Mitra,":[9],"LimiX,":[10],"and":[11,162,183,195,207],"TabDPT,":[12],"achieve":[13],"strong":[14],"zero-shot":[15],"performance":[16],"through":[17,129],"in-context":[18],"learning,":[19],"but":[20],"their":[21],"inductive":[22,118],"biases":[23,119],"remain":[24],"fixed":[25],"at":[26],"inference":[27,208],"time.":[28,209],"Adapting":[29],"a":[30,34,84,103,134],"pretrained":[31,122],"TFM":[32,99,144],"to":[33,60,96,111,141],"specific":[35],"dataset":[36],"or":[37,48,76],"task":[38],"typically":[39],"requires":[40],"either":[41],"full":[42],"fine-tuning,":[43],"which":[44,56],"is":[45,78,90,126,172],"computationally":[46],"expensive,":[47],"parameter-efficient":[49],"tuning":[50,182],"methods":[51],"(PEFT)":[52],"LoRA,":[55],"must":[57],"be":[58],"tailored":[59],"the":[61,68,97,108,113,117,121,130,142,166,173,177,191,198],"internal":[62],"architecture":[63],"of":[64,120,201],"each":[65],"TFM.":[66],"Furthermore,":[67],"evidence":[69],"on":[70,150,169,176,197],"whether":[71],"weight-space":[72],"fine-tuning":[73],"improves":[74],"accuracy":[75],"calibration":[77],"mixed":[79],"\\citep{tanna_exploring_2026,rubachev_finetuning_2025}.":[80],"We":[81],"introduce":[82],"TFM-Retouche,":[83],"lightweight":[85],"input-space":[86],"residual":[87,105],"adapter":[88,125],"that":[89,138],"architecture-agnostic":[91],"by":[92,188],"design":[93],"with":[94,116,133,179],"respect":[95],"frozen":[98,131,192],"backbone.":[100],"TFM-Retouche":[101],"learns":[102],"small":[104],"correction":[106],"in":[107],"input":[109,114],"space":[110],"align":[112],"data":[115],"model.":[123],"The":[124],"trained":[127],"end-to-end":[128],"TFM,":[132],"post-training":[135],"identity":[136],"guard":[137],"falls":[139],"back":[140],"unmodified":[143],"whenever":[145],"adaptation":[146],"does":[147],"not":[148],"help":[149],"held-out":[151],"validation.":[152],"On":[153],"TabArena-Lite":[154],"(51":[155],"datasets":[156],"spanning":[157],"binary":[158],"classification,":[159,161],"multiclass":[160],"regression),":[163],"TabICLv2-Retouche":[164],"--":[165,171],"framework":[167],"instantiated":[168],"TabICLv2":[170,193],"top-ranked":[174],"method":[175],"leaderboard":[178],"light":[180],"per-task":[181],"ensembling,":[184],"lifting":[185],"aggregate":[186],"Elo":[187],"+56":[189],"over":[190],"base":[194],"sitting":[196],"Pareto":[199],"front":[200],"predictive":[202],"quality":[203],"versus":[204],"both":[205],"training":[206]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-09T00:00:00"}