{"id":"https://openalex.org/W4416750073","doi":"https://doi.org/10.1109/iros60139.2025.11247608","title":"Multi-PrefDrive: Optimizing Large Language Models for Autonomous Driving Through Multi-Preference Tuning","display_name":"Multi-PrefDrive: Optimizing Large Language Models for Autonomous Driving Through Multi-Preference Tuning","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4416750073","doi":"https://doi.org/10.1109/iros60139.2025.11247608"},"language":null,"primary_location":{"id":"doi:10.1109/iros60139.2025.11247608","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11247608","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5071303232","display_name":"Yun Li","orcid":"https://orcid.org/0009-0002-7824-7751"},"institutions":[{"id":"https://openalex.org/I161296585","display_name":"Tokyo University of Science","ror":"https://ror.org/05sj3n476","country_code":"JP","type":"education","lineage":["https://openalex.org/I161296585"]},{"id":"https://openalex.org/I74801974","display_name":"The University of Tokyo","ror":"https://ror.org/057zh3y96","country_code":"JP","type":"education","lineage":["https://openalex.org/I74801974"]}],"countries":["JP"],"is_corresponding":true,"raw_author_name":"Yun Li","raw_affiliation_strings":["The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan","institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080705403","display_name":"Ehsan Javanmardi","orcid":"https://orcid.org/0000-0003-0337-115X"},"institutions":[{"id":"https://openalex.org/I74801974","display_name":"The University of Tokyo","ror":"https://ror.org/057zh3y96","country_code":"JP","type":"education","lineage":["https://openalex.org/I74801974"]},{"id":"https://openalex.org/I161296585","display_name":"Tokyo University of Science","ror":"https://ror.org/05sj3n476","country_code":"JP","type":"education","lineage":["https://openalex.org/I161296585"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Ehsan Javanmardi","raw_affiliation_strings":["The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan","institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108120651","display_name":"Simon Thompson","orcid":"https://orcid.org/0000-0003-0264-2579"},"institutions":[{"id":"https://openalex.org/I161296585","display_name":"Tokyo University of Science","ror":"https://ror.org/05sj3n476","country_code":"JP","type":"education","lineage":["https://openalex.org/I161296585"]},{"id":"https://openalex.org/I74801974","display_name":"The University of Tokyo","ror":"https://ror.org/057zh3y96","country_code":"JP","type":"education","lineage":["https://openalex.org/I74801974"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Simon Thompson","raw_affiliation_strings":["The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan","institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5107590712","display_name":"Kai Katsumata","orcid":null},"institutions":[{"id":"https://openalex.org/I74801974","display_name":"The University of Tokyo","ror":"https://ror.org/057zh3y96","country_code":"JP","type":"education","lineage":["https://openalex.org/I74801974"]},{"id":"https://openalex.org/I161296585","display_name":"Tokyo University of Science","ror":"https://ror.org/05sj3n476","country_code":"JP","type":"education","lineage":["https://openalex.org/I161296585"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Kai Katsumata","raw_affiliation_strings":["The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan","institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057553089","display_name":"Alex Orsholits","orcid":"https://orcid.org/0000-0001-8517-8753"},"institutions":[{"id":"https://openalex.org/I161296585","display_name":"Tokyo University of Science","ror":"https://ror.org/05sj3n476","country_code":"JP","type":"education","lineage":["https://openalex.org/I161296585"]},{"id":"https://openalex.org/I74801974","display_name":"The University of Tokyo","ror":"https://ror.org/057zh3y96","country_code":"JP","type":"education","lineage":["https://openalex.org/I74801974"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Alex Orsholits","raw_affiliation_strings":["The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan","institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5067716610","display_name":"Manabu Tsukada","orcid":"https://orcid.org/0000-0001-8045-3939"},"institutions":[{"id":"https://openalex.org/I74801974","display_name":"The University of Tokyo","ror":"https://ror.org/057zh3y96","country_code":"JP","type":"education","lineage":["https://openalex.org/I74801974"]},{"id":"https://openalex.org/I161296585","display_name":"Tokyo University of Science","ror":"https://ror.org/05sj3n476","country_code":"JP","type":"education","lineage":["https://openalex.org/I161296585"]}],"countries":["JP"],"is_corresponding":false,"raw_author_name":"Manabu Tsukada","raw_affiliation_strings":["The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan"],"affiliations":[{"raw_affiliation_string":"The University of Tokyo,Graduate School of Information and Science Technology,Tokyo,Japan","institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5071303232"],"corresponding_institution_ids":["https://openalex.org/I161296585","https://openalex.org/I74801974"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.42772401,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4347","last_page":"4354"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.23829999566078186,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.23829999566078186,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1915999948978424,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.14429999887943268,"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/margin","display_name":"Margin (machine learning)","score":0.5115000009536743},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.5041999816894531},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.4934999942779541},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.4551999866962433},{"id":"https://openalex.org/keywords/binary-number","display_name":"Binary number","score":0.4456999897956848},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.4244000017642975},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.4041999876499176}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6654999852180481},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.5115000009536743},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5041999816894531},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.4934999942779541},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.4551999866962433},{"id":"https://openalex.org/C48372109","wikidata":"https://www.wikidata.org/wiki/Q3913","display_name":"Binary number","level":2,"score":0.4456999897956848},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4390000104904175},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.4244000017642975},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.4041999876499176},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.399399995803833},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.38420000672340393},{"id":"https://openalex.org/C64543145","wikidata":"https://www.wikidata.org/wiki/Q162942","display_name":"Intersection (aeronautics)","level":2,"score":0.3452000021934509},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3337000012397766},{"id":"https://openalex.org/C2780689630","wikidata":"https://www.wikidata.org/wiki/Q2081815","display_name":"Driving simulator","level":2,"score":0.33250001072883606},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.28700000047683716},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.27309998869895935},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.27160000801086426},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2676999866962433},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.26660001277923584},{"id":"https://openalex.org/C63435697","wikidata":"https://www.wikidata.org/wiki/Q864135","display_name":"Binary code","level":3,"score":0.25440001487731934},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2533000111579895}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iros60139.2025.11247608","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11247608","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":11,"referenced_works":["https://openalex.org/W2968008415","https://openalex.org/W2968296999","https://openalex.org/W3037429136","https://openalex.org/W3210290940","https://openalex.org/W4388819818","https://openalex.org/W4394862623","https://openalex.org/W4401386967","https://openalex.org/W4402592483","https://openalex.org/W4402727495","https://openalex.org/W4408698480","https://openalex.org/W4412886810"],"related_works":[],"abstract_inverted_index":{"This":[0],"paper":[1],"introduces":[2],"Multi-PrefDrive,":[3],"a":[4,39,137],"framework":[5],"that":[6,91,126],"significantly":[7],"enhances":[8],"LLM-based":[9,181],"autonomous":[10,182],"driving":[11,20,27,65,151,183],"through":[12],"multidimensional":[13],"preference":[14,45,71],"tuning.":[15],"Aligning":[16],"LLMs":[17],"with":[18,58,112],"human":[19],"preferences":[21],"is":[22],"crucial":[23],"yet":[24],"challenging,":[25],"as":[26],"scenarios":[28],"involve":[29],"complex":[30],"decisions":[31],"where":[32],"multiple":[33,59],"incorrect":[34],"actions":[35,78],"can":[36],"correspond":[37],"to":[38,48,149,162,177],"single":[40],"correct":[41],"choice.":[42],"Traditional":[43],"binary":[44],"tuning":[46],"fails":[47],"capture":[49],"this":[50],"complexity.":[51],"Our":[52],"approach":[53],"pairs":[54],"each":[55],"chosen":[56,131],"action":[57],"rejected":[60,133],"alternatives,":[61],"better":[62],"reflecting":[63],"real-world":[64],"decisions.":[66],"By":[67],"implementing":[68],"the":[69,80,87],"Plackett-Luce":[70],"model,":[72],"we":[73],"enable":[74,163],"nuanced":[75],"ranking":[76],"of":[77,82,140],"across":[79],"spectrum":[81],"possible":[83],"errors.":[84],"Experiments":[85],"in":[86,98,105,115,180],"CARLA":[88],"simulator":[89],"demonstrate":[90],"our":[92,171],"algorithm":[93],"achieves":[94],"an":[95,102],"11.0%":[96],"improvement":[97],"overall":[99],"score":[100],"and":[101,122,132,142,159,168,174],"83.6%":[103],"reduction":[104],"infrastructure":[106],"collisions,":[107],"while":[108],"showing":[109],"perfect":[110],"compliance":[111],"traffic":[113],"signals":[114],"certain":[116],"environments.":[117],"Comparative":[118],"analysis":[119],"against":[120],"DPO":[121],"its":[123],"variants":[124],"reveals":[125],"Multi-PrefDrive\u2019s":[127],"superior":[128],"discrimination":[129],"between":[130],"actions,":[134],"which":[135],"achieving":[136],"margin":[138],"value":[139],"25,":[141],"such":[143],"ability":[144],"has":[145],"been":[146],"directly":[147],"translates":[148],"enhanced":[150],"performance.":[152],"We":[153],"implement":[154],"memory-efficient":[155],"techniques":[156],"including":[157],"LoRA":[158],"4-bit":[160],"quantization":[161],"deployment":[164],"on":[165],"consumer-grade":[166],"hardware":[167],"will":[169],"open-source":[170],"training":[172],"code":[173],"multi-rejected":[175],"dataset":[176],"advance":[178],"research":[179],"systems.":[184],"Project":[185],"Page":[186],"(https://liyun0607.github.io/).":[187]},"counts_by_year":[],"updated_date":"2026-03-07T16:01:11.037858","created_date":"2025-11-28T00:00:00"}