{"id":"https://openalex.org/W7159628509","doi":"https://doi.org/10.48550/arxiv.2604.27505","title":"Leveraging Verifier-Based Reinforcement Learning in Image Editing","display_name":"Leveraging Verifier-Based Reinforcement Learning in Image Editing","publication_year":2026,"publication_date":"2026-04-30","ids":{"openalex":"https://openalex.org/W7159628509","doi":"https://doi.org/10.48550/arxiv.2604.27505"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.27505","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.27505","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.2604.27505","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134965157","display_name":"Hanzhong Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Hanzhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134963296","display_name":"Jie Wu (10038)","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134929672","display_name":"Jie Liu (15128)","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134935853","display_name":"Yu Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134934679","display_name":"Zilyu Ye","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye, Zilyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079045180","display_name":"LinXiao Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Linxiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091579107","display_name":"Xionghui Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xionghui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134975002","display_name":"Yizhou Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Yizhou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134975507","display_name":"Weilin Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Weilin","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":0.36570000648498535,"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":0.36570000648498535,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.13850000500679016,"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/T12377","display_name":"Digital Humanities and Scholarship","score":0.06650000065565109,"subfield":{"id":"https://openalex.org/subfields/1208","display_name":"Literature and Literary Theory"},"field":{"id":"https://openalex.org/fields/12","display_name":"Arts and Humanities"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7982000112533569},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.654699981212616},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5655999779701233},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.5138000249862671},{"id":"https://openalex.org/keywords/image-editing","display_name":"Image editing","score":0.5092999935150146},{"id":"https://openalex.org/keywords/pairwise-comparison","display_name":"Pairwise comparison","score":0.47519999742507935},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.4059999883174896}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8366000056266785},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7982000112533569},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.654699981212616},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6122000217437744},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5655999779701233},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.5138000249862671},{"id":"https://openalex.org/C2776674983","wikidata":"https://www.wikidata.org/wiki/Q545981","display_name":"Image editing","level":3,"score":0.5092999935150146},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.47519999742507935},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47049999237060547},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.4059999883174896},{"id":"https://openalex.org/C2777984123","wikidata":"https://www.wikidata.org/wiki/Q9248237","display_name":"Pointwise","level":2,"score":0.39239999651908875},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.3644999861717224},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3149000108242035},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.31220000982284546},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.267300009727478},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.25360000133514404}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.27505","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.27505","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.2604.27505","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.27505","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"While":[0],"Reinforcement":[1],"Learning":[2],"from":[3,69,213],"Human":[4],"Feedback":[5],"(RLHF)":[6],"has":[7],"become":[8],"a":[9,30,70,74,80,84,135,150,205],"pivotal":[10],"paradigm":[11],"for":[12,35,96],"text-to-image":[13],"generation,":[14],"its":[15,228],"application":[16],"to":[17,67,73,137,160,172,215,222],"image":[18,98,110,232],"editing":[19,37,174,223],"remains":[20],"largely":[21],"unexplored.":[22],"A":[23],"key":[24,65],"bottleneck":[25],"is":[26,66],"the":[27,64,108,167],"lack":[28],"of":[29],"robust":[31],"general":[32],"reward":[33,41,89,140,181,200],"model":[34,90],"all":[36],"tasks.":[38],"Existing":[39],"edit":[40],"models":[42,175,224],"usually":[43],"give":[44],"overall":[45],"scores":[46],"without":[47],"detailed":[48],"checks,":[49],"ignoring":[50],"different":[51],"instruction":[52],"requirements":[53],"and":[54,92,114,195,202],"causing":[55],"biased":[56],"rewards.":[57],"To":[58,123],"address":[59],"this,":[60],"we":[61,128,143,169,203],"argue":[62],"that":[63,82,154,186],"move":[68],"simple":[71],"scorer":[72],"reasoning":[75,88],"verifier.":[76],"We":[77],"introduce":[78,144],"Edit-R1,":[79],"framework":[81],"builds":[83],"chain-of-thought":[85],"(CoT)":[86],"verifier-based":[87],"(RRM)":[91],"then":[93],"leverages":[94,155],"it":[95],"downstream":[97],"editing.":[99,233],"The":[100],"Edit-RRM":[101,188],"breaks":[102],"instructions":[103],"into":[104,118],"distinct":[105],"principles,":[106],"evaluates":[107],"edited":[109],"against":[111],"each":[112],"principle,":[113],"aggregates":[115],"these":[116],"checks":[117],"an":[119,126,198],"interpretable,":[120],"fine-grained":[121],"reward.":[122],"build":[124],"such":[125,192],"RRM,":[127,168],"first":[129],"apply":[130],"supervised":[131],"fine-tuning":[132],"(SFT)":[133],"as":[134,193,197],"``cold-start''":[136],"generate":[138],"CoT":[139],"trajectories.":[141],"Then,":[142],"Group":[145],"Contrastive":[146],"Preference":[147],"Optimization":[148],"(GCPO),":[149],"reinforcement":[151],"learning":[152],"algorithm":[153],"human":[156],"pairwise":[157],"preference":[158],"data":[159],"reinforce":[161],"our":[162,187],"pointwise":[163],"RRM.":[164],"After":[165],"building":[166],"use":[170],"GRPO":[171],"train":[173],"with":[176,209],"this":[177],"non-differentiable":[178],"yet":[179],"powerful":[180,190],"model.":[182],"Extensive":[183],"experiments":[184],"demonstrate":[185],"surpasses":[189],"VLMs":[191],"Seed-1.5-VL":[194],"Seed-1.6-VL":[196],"editing-specific":[199],"model,":[201],"observe":[204],"clear":[206],"scaling":[207],"trend,":[208],"performance":[210],"consistently":[211],"improving":[212],"3B":[214],"7B":[216],"parameters.":[217],"Moreover,":[218],"Edit-R1":[219],"delivers":[220],"gains":[221],"like":[225],"FLUX.1-kontext,":[226],"highlighting":[227],"effectiveness":[229],"in":[230],"enhancing":[231]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-02T00:00:00"}