{"id":"https://openalex.org/W7140143045","doi":"https://doi.org/10.48550/arxiv.2603.19779","title":"One Model, Two Minds: Task-Conditioned Reasoning for Unified Image Quality and Aesthetic Assessment","display_name":"One Model, Two Minds: Task-Conditioned Reasoning for Unified Image Quality and Aesthetic Assessment","publication_year":2026,"publication_date":"2026-03-20","ids":{"openalex":"https://openalex.org/W7140143045","doi":"https://doi.org/10.48550/arxiv.2603.19779"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.19779","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19779","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":null,"license_id":null,"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.2603.19779","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5058029288","display_name":"Wen Yin","orcid":"https://orcid.org/0000-0002-7367-1820"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yin, Wen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130372412","display_name":"Cencen Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Cencen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130351586","display_name":"Dingrui Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Dingrui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022837060","display_name":"Bing Su","orcid":"https://orcid.org/0000-0002-4379-9014"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Su, Bing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017943466","display_name":"Yuan-Fang Li","orcid":"https://orcid.org/0000-0003-4651-2821"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Yuan-Fang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130379463","display_name":"Tao He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Tao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.6527000069618225,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.6527000069618225,"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/T12650","display_name":"Aesthetic Perception and Analysis","score":0.31349998712539673,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T11165","display_name":"Image and Video Quality Assessment","score":0.0038999998942017555,"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/quality","display_name":"Quality (philosophy)","score":0.6144000291824341},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.6032999753952026},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.5383999943733215},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.47760000824928284},{"id":"https://openalex.org/keywords/tatar","display_name":"Tatar","score":0.4194999933242798},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.384799987077713},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.37209999561309814}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6516000032424927},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6233000159263611},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.6144000291824341},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.6032999753952026},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.5383999943733215},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.47760000824928284},{"id":"https://openalex.org/C2780124745","wikidata":"https://www.wikidata.org/wiki/Q25285","display_name":"Tatar","level":2,"score":0.4194999933242798},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3939000070095062},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.384799987077713},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.37209999561309814},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3573000133037567},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.31850001215934753},{"id":"https://openalex.org/C55020928","wikidata":"https://www.wikidata.org/wiki/Q3813865","display_name":"Image quality","level":3,"score":0.3050999939441681},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.2700999975204468},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.25769999623298645},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2556000053882599},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.19779","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19779","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.19779","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.19779","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"score":0.4071499705314636,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Unifying":[0],"Image":[1,6],"Quality":[2],"Assessment":[3,8],"(IQA)":[4],"and":[5,32,50,62,77,81,132,149,159,183,192],"Aesthetic":[7],"(IAA)":[9],"in":[10],"a":[11,23,74,102,207],"single":[12],"multimodal":[13],"large":[14],"language":[15],"model":[16],"is":[17,40,63,216],"appealing,":[18],"yet":[19],"existing":[20],"methods":[21],"adopt":[22],"task-agnostic":[24],"recipe":[25],"that":[26,105,125,141,152,171],"applies":[27],"the":[28,107],"same":[29],"reasoning":[30,75,123],"strategy":[31],"reward":[33],"to":[34],"both":[35,86,179],"tasks.":[36],"We":[37,70],"show":[38],"this":[39],"fundamentally":[41],"misaligned:":[42],"IQA":[43,127,158],"relies":[44],"on":[45,113,178],"low-level,":[46],"objective":[47],"perceptual":[48,130,212],"cues":[49],"benefits":[51],"from":[52],"concise":[53,129],"distortion-focused":[54],"reasoning,":[55],"whereas":[56],"IAA":[57,133],"requires":[58],"deliberative":[59,135],"semantic":[60],"judgment":[61],"poorly":[64],"served":[65],"by":[66,91],"point-wise":[67],"score":[68,155],"regression.":[69],"identify":[71],"these":[72,92],"as":[73,206],"mismatch":[76],"an":[78],"optimization":[79],"mismatch,":[80],"provide":[82],"empirical":[83],"evidence":[84],"for":[85,157,163,198,210],"through":[87],"controlled":[88],"probes.":[89],"Motivated":[90],"findings,":[93],"we":[94],"propose":[95],"TATAR":[96,117,172],"(Task-Aware":[97],"Thinking":[98],"with":[99,128,134,188],"Asymmetric":[100],"Rewards),":[101],"unified":[103,176,211],"framework":[104],"shares":[106],"visual-language":[108],"backbone":[109],"while":[110],"conditioning":[111],"post-training":[112,205],"each":[114],"task's":[115],"nature.":[116],"combines":[118],"three":[119],"components:":[120],"fast--slow":[121],"task-specific":[122,189],"construction":[124],"pairs":[126],"rationales":[131],"aesthetic":[136,199],"narratives;":[137],"two-stage":[138],"SFT+GRPO":[139],"learning":[140],"establishes":[142],"task-aware":[143],"behavioral":[144],"priors":[145],"before":[146],"reward-driven":[147],"refinement;":[148],"asymmetric":[150],"rewards":[151],"apply":[153],"Gaussian":[154],"shaping":[156],"Thurstone-style":[160],"completion":[161],"ranking":[162],"IAA.":[164],"Extensive":[165],"experiments":[166],"across":[167],"eight":[168],"benchmarks":[169],"demonstrate":[170],"consistently":[173],"outperforms":[174],"prior":[175],"baselines":[177],"tasks":[180],"under":[181],"in-domain":[182],"cross-domain":[184],"settings,":[185],"remains":[186],"competitive":[187],"specialized":[190],"models,":[191],"yields":[193],"more":[194],"stable":[195],"training":[196],"dynamics":[197],"assessment.":[200],"Our":[201,214],"results":[202],"establish":[203],"task-conditioned":[204],"principled":[208],"paradigm":[209],"scoring.":[213],"code":[215],"publicly":[217],"available":[218],"at":[219],"https://github.com/yinwen2019/TATAR.":[220]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-24T00:00:00"}