{"id":"https://openalex.org/W7203937564","doi":"https://doi.org/10.48550/arxiv.2608.19598","title":"PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignment","display_name":"PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignment","publication_year":2026,"publication_date":"2026-08-20","ids":{"openalex":"https://openalex.org/W7203937564","doi":"https://doi.org/10.48550/arxiv.2608.19598"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.19598","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.19598","pdf_url":"https://arxiv.org/pdf/2608.19598","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","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.19598","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5147701992","display_name":"Jiawei Feng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Feng, Jiawei","raw_affiliation_strings":["University of Science and Technology of China Hefei , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China Hefei , China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075695432","display_name":"Jiancan Wu","orcid":"https://orcid.org/0000-0002-6941-5218","observed_orcids":["https://orcid.org/0000-0002-6941-5218"]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Jiancan","raw_affiliation_strings":["University of Science and Technology of China Hefei , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China Hefei , China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147808146","display_name":"Xingyu Zhu","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":"Zhu, Xingyu","raw_affiliation_strings":["National University of Singapore Kent Ridge , Singapore"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"National University of Singapore Kent Ridge , Singapore","institution_ids":["https://openalex.org/I165932596"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5151846710","display_name":"Junkang Wu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Junkang","raw_affiliation_strings":["University of Science and Technology of China Hefei , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China Hefei , China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147809247","display_name":"Xiang Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Xiang","raw_affiliation_strings":["University of Science and Technology of China Hefei , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China Hefei , China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5127314158","display_name":"Xiangnan He","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"He, Xiangnan","raw_affiliation_strings":["University of Science and Technology of China Hefei , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Science and Technology of China Hefei , China","institution_ids":["https://openalex.org/I126520041"]}]}],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9980000257492065,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9980000257492065,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.00039999998989515007,"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":0.0003000000142492354,"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/direct-preference-optimization","display_name":"direct preference optimization","score":0.8360000252723694},{"id":"https://openalex.org/keywords/multimodal-large-language-models","display_name":"multimodal large language models","score":0.9459999799728394},{"id":"https://openalex.org/keywords/preference-alignment","display_name":"preference alignment","score":0.5149999856948853},{"id":"https://openalex.org/keywords/hallucination-mitigation","display_name":"hallucination mitigation","score":0.7549999952316284},{"id":"https://openalex.org/keywords/image-text-alignment","display_name":"image-text alignment","score":0.5019999742507935},{"id":"https://openalex.org/keywords/language-model-fine-tuning","display_name":"language model fine-tuning","score":0.5509999990463257}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.724399983882904},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.641700029373169},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.636900007724762},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6182000041007996},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.48669999837875366},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4839000105857849},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.45680001378059387},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.4341000020503998},{"id":"https://openalex.org/C2780878386","wikidata":"https://www.wikidata.org/wiki/Q1659648","display_name":"Visual language","level":2,"score":0.42730000615119934},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.38449999690055847},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.36980000138282776},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.32820001244544983},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.32350000739097595},{"id":"https://openalex.org/C64754055","wikidata":"https://www.wikidata.org/wiki/Q7574053","display_name":"Spatial contextual awareness","level":2,"score":0.31520000100135803},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.3100999891757965},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.30959999561309814},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.29350000619888306},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2759999930858612},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.2743000090122223},{"id":"https://openalex.org/C2777868144","wikidata":"https://www.wikidata.org/wiki/Q7239817","display_name":"Preference elicitation","level":3,"score":0.2694000005722046},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.2662000060081482},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2651999890804291},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2637999951839447},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.2533999979496002},{"id":"https://openalex.org/C112313634","wikidata":"https://www.wikidata.org/wiki/Q7886648","display_name":"Complement (music)","level":5,"score":0.25130000710487366}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.19598","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.19598","pdf_url":"https://arxiv.org/pdf/2608.19598","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.2608.19598","doi":"https://doi.org/10.48550/arxiv.2608.19598","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.19598","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:2608.19598","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.19598","pdf_url":"https://arxiv.org/pdf/2608.19598","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/W7203937564.pdf","grobid_xml":"https://content.openalex.org/works/W7203937564.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7153781235","https://openalex.org/W4404782590","https://openalex.org/W4414631762","https://openalex.org/W7119105704","https://openalex.org/W7215078982","https://openalex.org/W4400374599","https://openalex.org/W7140198454","https://openalex.org/W7160515398","https://openalex.org/W7167799582","https://openalex.org/W4406950799"],"abstract_inverted_index":{"Direct":[0],"Preference":[1],"Optimization":[2],"(DPO)":[3],"has":[4],"emerged":[5],"as":[6],"an":[7],"effective":[8],"approach":[9],"for":[10,86],"aligning":[11],"large":[12],"language":[13,127],"models":[14,44],"(LLMs)":[15],"with":[16,53],"human":[17],"preferences.":[18],"However,":[19],"its":[20],"adaptation":[21],"to":[22,47,96,121],"multimodal":[23,36,87,153],"settings":[24],"remains":[25],"unexplored.":[26],"Through":[27],"representational":[28],"analysis,":[29],"we":[30,40,78],"identify":[31],"a":[32,84,103],"key":[33],"limitation":[34],"in":[35],"preference":[37,94],"optimization,":[38],"which":[39,90],"term":[41],"visual":[42,55,93,98,122,149],"insensitivity:":[43],"often":[45],"fail":[46],"distinguish":[48],"between":[49],"images":[50],"and":[51,71,155],"those":[52],"critical":[54],"context":[56,123],"removed.":[57],"Our":[58],"theoretical":[59,104],"analysis":[60,105],"further":[61,101],"uncovers":[62],"two":[63],"manifestations":[64],"of":[65,130,141],"this":[66],"problem,":[67],"namely":[68],"Across-Image":[69],"Insensitivity":[70],"Within-Image":[72],"Insensitivity.":[73],"To":[74],"address":[75],"these":[76],"challenges,":[77],"propose":[79],"Perception-Enhanced":[80],"Alignment":[81],"DPO":[82],"(PEA-DPO),":[83],"framework":[85],"LLMs":[88],"alignment,":[89,154],"explicitly":[91],"leverages":[92],"signals":[95],"overcome":[97],"insensitivity.":[99],"We":[100],"provide":[102],"demonstrating":[106],"that":[107,117,145],"PEA-DPO":[108,118,146],"provably":[109],"mitigates":[110,148],"both":[111],"failure":[112],"modes.":[113],"Empirical":[114],"results":[115],"demonstrate":[116],"enhances":[119],"sensitivity":[120],"while":[124],"preserving":[125],"the":[126,131],"modeling":[128],"capacity":[129],"base":[132],"model.":[133],"Evaluations":[134],"across":[135],"three":[136],"hallucination":[137],"benchmarks":[138],"using":[139],"MLLMs":[140],"varying":[142],"scales":[143],"show":[144],"effectively":[147],"insensitivity,":[150],"achieves":[151],"stronger":[152],"substantially":[156],"reduces":[157],"hallucinations.":[158]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-22T00:00:00"}