{"id":"https://openalex.org/W7214707673","doi":"https://doi.org/10.48550/arxiv.2609.35109","title":"Using Context Is Not Enough: Test-Time Training for Personalized Reward Modeling","display_name":"Using Context Is Not Enough: Test-Time Training for Personalized Reward Modeling","publication_year":2026,"publication_date":"2026-09-28","ids":{"openalex":"https://openalex.org/W7214707673","doi":"https://doi.org/10.48550/arxiv.2609.35109"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.35109","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.35109","pdf_url":"https://arxiv.org/pdf/2609.35109","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":null,"license_id":null,"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/2609.35109","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153552047","display_name":"Bohao Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Bohao","raw_affiliation_strings":["Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153435116","display_name":"Xiaoyan Zhao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zhao, Xiaoyan","raw_affiliation_strings":["National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153454099","display_name":"Yang Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Zhang, Yang","raw_affiliation_strings":["National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153505765","display_name":"Jinghang Guo","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I165932596","display_name":"National University of Singapore","ror":"https://ror.org/01tgyzw49","country_code":"SG","type":"education","lineage":["https://openalex.org/I165932596"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Guo, Jinghang","raw_affiliation_strings":["National University of Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153466111","display_name":"Chun Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Chun","raw_affiliation_strings":["Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153529604","display_name":"Can Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Can","raw_affiliation_strings":["Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5153456236","display_name":"Jiawei Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Jiawei","raw_affiliation_strings":["Zhejiang University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Zhejiang University","institution_ids":["https://openalex.org/I76130692"]}]}],"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":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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.3822999894618988,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.3822999894618988,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.12610000371932983,"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/T10028","display_name":"Topic Modeling","score":0.09520000219345093,"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/reinforcement-learning-from-human-feedback","display_name":"reinforcement learning from human feedback","score":0.9020000100135803},{"id":"https://openalex.org/keywords/test-time-training","display_name":"test-time training","score":0.8529999852180481},{"id":"https://openalex.org/keywords/in-context-learning","display_name":"in-context learning","score":0.843999981880188},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7730000019073486},{"id":"https://openalex.org/keywords/preference-alignment","display_name":"preference alignment","score":0.5649999976158142}],"concepts":[{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.7477999925613403},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7347999811172485},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.6517000198364258},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6449000239372253},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6182000041007996},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.6139000058174133},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.5916000008583069},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5519999861717224},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5386000275611877},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5245000123977661},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3393000066280365},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.33559998869895935},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.32010000944137573},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3140000104904175},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.2574999928474426}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.35109","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.35109","pdf_url":"https://arxiv.org/pdf/2609.35109","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.35109","doi":"https://doi.org/10.48550/arxiv.2609.35109","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.35109","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.35109","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.35109","pdf_url":"https://arxiv.org/pdf/2609.35109","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":null,"license_id":null,"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/W7214707673.pdf","grobid_xml":"https://content.openalex.org/works/W7214707673.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W4400435102","https://openalex.org/W4401306886","https://openalex.org/W4403796339","https://openalex.org/W4404344563","https://openalex.org/W6910665759","https://openalex.org/W6948294992","https://openalex.org/W6966976602","https://openalex.org/W7131426165","https://openalex.org/W7137100030","https://openalex.org/W7165646981","https://openalex.org/W7167932541","https://openalex.org/W7172397141"],"related_works":["https://openalex.org/W4416036158","https://openalex.org/W4403995389","https://openalex.org/W6948294992","https://openalex.org/W4417159287","https://openalex.org/W6891914921","https://openalex.org/W4415238136","https://openalex.org/W7128769370","https://openalex.org/W4391709257","https://openalex.org/W4386554585","https://openalex.org/W7102415536"],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1,44],"from":[2],"human":[3,12],"feedback":[4],"(RLHF)":[5],"aligns":[6],"large":[7,166],"language":[8],"models":[9,30],"(LLMs)":[10],"with":[11,115],"preferences,":[13],"yet":[14],"most":[15,40],"pipelines":[16],"learn":[17],"a":[18,47,60,116,142,165],"single":[19,143],"reward":[20,29,97],"model":[21],"that":[22,119,152],"overlooks":[23],"individual":[24],"differences":[25],"in":[26],"preferences.":[27],"Personalized":[28],"(PRMs)":[31],"address":[32,78],"this":[33],"by":[34,74,164],"conditioning":[35],"rewards":[36],"on":[37],"user-specific":[38,92],"feedback,":[39,113],"commonly":[41],"through":[42],"in-context":[43],"(ICL),":[45],"where":[46],"user's":[48],"historical":[49,157],"comparisons":[50],"are":[51],"supplied":[52],"as":[53],"contextual":[54,75],"preference":[55,71,112,123,158],"pairs.":[56,76],"However,":[57],"we":[58,80],"identify":[59],"key":[61],"limitation":[62],"of":[63,111],"ICL-based":[64],"PRMs:":[65],"they":[66],"fail":[67],"to":[68,106,125,133],"capture":[69],"the":[70,108],"relations":[72,90,124,159],"conveyed":[73],"To":[77],"this,":[79],"propose":[81],"Preference-Aligned":[82],"Test-Time":[83],"Training":[84],"(P-TTT),":[85],"which":[86],"explicitly":[87],"encodes":[88],"these":[89],"into":[91],"fast":[93,139],"weights":[94,140],"for":[95],"personalized":[96],"prediction.":[98],"P-TTT":[99,130,153],"introduces":[100],"sequence-level":[101],"update":[102],"and":[103,135,160],"apply":[104],"operations":[105],"match":[107],"response-level":[109],"granularity":[110],"together":[114],"preference-aligned":[117],"objective":[118],"directly":[120],"uses":[121],"pairwise":[122],"guide":[126],"fast-weight":[127],"adaptation.":[128],"Notably,":[129],"is":[131],"simple":[132],"implement":[134],"computationally":[136],"efficient,":[137],"updating":[138],"within":[141],"forward":[144],"pass":[145],"without":[146],"inference-time":[147],"backpropagation.":[148],"Extensive":[149],"experiments":[150],"show":[151],"more":[154],"effectively":[155],"captures":[156],"outperforms":[161],"state-of-the-art":[162],"methods":[163],"margin.":[167]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-30T00:00:00"}