{"id":"https://openalex.org/W7161123735","doi":"https://doi.org/10.48550/arxiv.2605.12724","title":"Inline Critic Steers Image Editing","display_name":"Inline Critic Steers Image Editing","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161123735","doi":"https://doi.org/10.48550/arxiv.2605.12724"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.12724","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12724","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.12724","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113395507","display_name":"Weitai Kang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kang, Weitai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040988693","display_name":"Xiaohang Zhan","orcid":"https://orcid.org/0000-0003-2136-7592"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhan, Xiaohang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136108926","display_name":"Yizhou Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yizhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063915104","display_name":"Mang Tik Chiu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chiu, Mang Tik","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136168624","display_name":"Jason Kuen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kuen, Jason","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136181937","display_name":"Kangning Liu","orcid":"https://orcid.org/0009-0000-0398-9879"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Kangning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136149838","display_name":"Yan Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9004999995231628,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.9004999995231628,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.009200000204145908,"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/T12859","display_name":"Cell Image Analysis Techniques","score":0.008700000122189522,"subfield":{"id":"https://openalex.org/subfields/1304","display_name":"Biophysics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.6370999813079834},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.589900016784668},{"id":"https://openalex.org/keywords/ask-price","display_name":"Ask price","score":0.5501999855041504},{"id":"https://openalex.org/keywords/image-editing","display_name":"Image editing","score":0.5128999948501587},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.4674000144004822},{"id":"https://openalex.org/keywords/signal","display_name":"SIGNAL (programming language)","score":0.4277999997138977},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.4163999855518341}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6984999775886536},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.6370999813079834},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5992000102996826},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.589900016784668},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.5501999855041504},{"id":"https://openalex.org/C2776674983","wikidata":"https://www.wikidata.org/wiki/Q545981","display_name":"Image editing","level":3,"score":0.5128999948501587},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5008999705314636},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.4674000144004822},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.4277999997138977},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.4163999855518341},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.399399995803833},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.37290000915527344},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.3452000021934509},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.3294000029563904},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.31859999895095825},{"id":"https://openalex.org/C2987933465","wikidata":"https://www.wikidata.org/wiki/Q141130","display_name":"Image manipulation","level":3,"score":0.3100000023841858},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2930000126361847},{"id":"https://openalex.org/C146044194","wikidata":"https://www.wikidata.org/wiki/Q5157334","display_name":"Computational photography","level":4,"score":0.2879999876022339}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.12724","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12724","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.12724","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12724","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":"Preprint"},"sustainable_development_goals":[{"score":0.5678888559341431,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Instruction-based":[0],"image":[1,37],"editing":[2],"exhibits":[3],"heterogeneous":[4],"difficulty":[5],"not":[6],"only":[7,72],"across":[8,12],"cases":[9],"but":[10],"also":[11],"regions":[13],"of":[14,152],"an":[15,52],"image,":[16],"motivating":[17],"refinement":[18,29],"approaches":[19],"that":[20,67,107,182],"allocate":[21],"correction":[22],"to":[23,122,134,141,143],"where":[24],"the":[25,74,78,93,126,136,153,164,168,183,187],"model":[26,64],"struggles.":[27],"Existing":[28],"signals":[30],"arrive":[31],"late,":[32],"after":[33],"a":[34,39,47,61,104,109,147,158],"fully":[35],"generated":[36],"or":[38],"completed":[40],"denoising":[41],"step.":[42],"We":[43,177],"ask":[44],"whether":[45],"such":[46],"signal":[48],"can":[49],"act":[50],"within":[51],"ongoing":[53],"forward":[54,127],"pass.":[55,128],"To":[56],"investigate":[57],"this,":[58,99],"we":[59,100,149],"probe":[60],"frozen":[62,110],"image-editing":[63],"and":[65,117,167,190],"find":[66],"although":[68],"generation":[69,124],"capability":[70],"emerges":[71],"in":[73,84],"last":[75],"few":[76],"layers,":[77],"error":[79,95],"pattern":[80],"is":[81,132],"already":[82],"set":[83],"early":[85],"layers":[86,116],"(rank":[87],"correlation":[88],"\\r{ho}":[89],"=":[90],"0.83":[91],"with":[92],"final-layer":[94],"map).":[96],"Based":[97],"on":[98,155,161,172],"introduce":[101],"Inline":[102],"Critic,":[103],"learnable":[105],"token":[106],"critiques":[108],"model's":[111,188],"predictions":[112],"at":[113,193],"its":[114,119],"intermediate":[115],"steers":[118],"hidden":[120],"states":[121],"refine":[123],"during":[125],"A":[129],"three-stage":[130],"recipe":[131],"proposed":[133],"stabilize":[135],"training":[137],"from":[138],"learning":[139],"how":[140],"critique":[142],"steering":[144],"generation.":[145],"As":[146],"result,":[148],"achieve":[150],"state":[151],"art":[154],"GEdit-Bench":[156],"(7.89),":[157],"+9.4":[159],"gain":[160],"RISEBench":[162],"over":[163],"same":[165],"backbone,":[166],"strongest":[169],"open-source":[170],"result":[171],"KRIS-Bench":[173],"(81.92,":[174],"surpassing":[175],"GPT-4o).":[176],"further":[178],"provide":[179],"analyses":[180],"showing":[181],"critic":[184],"genuinely":[185],"shapes":[186],"attention":[189],"prediction":[191],"updates":[192],"subsequent":[194],"layers.":[195]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-15T00:00:00"}