{"id":"https://openalex.org/W7133336087","doi":"https://doi.org/10.48550/arxiv.2603.00141","title":"From Scale to Speed: Adaptive Test-Time Scaling for Image Editing","display_name":"From Scale to Speed: Adaptive Test-Time Scaling for Image Editing","publication_year":2026,"publication_date":"2026-02-24","ids":{"openalex":"https://openalex.org/W7133336087","doi":"https://doi.org/10.48550/arxiv.2603.00141"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.00141","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00141","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.2603.00141","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5069741268","display_name":"Xiangyan Qu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Qu, Xiangyan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127892320","display_name":"Zhenlong Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Zhenlong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128016671","display_name":"Jing Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Jing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127916543","display_name":"Rui Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Rui","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127963096","display_name":"Datao Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Datao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127968716","display_name":"Meng Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Meng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127893013","display_name":"Lei Sun (45752)","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Lei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127937965","display_name":"Yancheng Bai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bai, Yancheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127968298","display_name":"Xiangxiang Chu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chu, Xiangxiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121141750","display_name":"Gaopeng Gou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gou, Gaopeng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127899139","display_name":"Gang Xiong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiong, Gang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5127901531","display_name":"Yujun Cai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cai, Yujun","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":12,"corresponding_author_ids":["https://openalex.org/A5069741268"],"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":0.8310999870300293,"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":0.8310999870300293,"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":0.062300000339746475,"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":0.009399999864399433,"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/image-editing","display_name":"Image editing","score":0.6660000085830688},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.578000009059906},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.5652999877929688},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5647000074386597},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5443000197410583},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.5277000069618225},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5065000057220459},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.48350000381469727}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7882000207901001},{"id":"https://openalex.org/C2776674983","wikidata":"https://www.wikidata.org/wiki/Q545981","display_name":"Image editing","level":3,"score":0.6660000085830688},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.578000009059906},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.5652999877929688},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5647000074386597},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5443000197410583},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.5277000069618225},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5065000057220459},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.48350000381469727},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4756999909877777},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4528999924659729},{"id":"https://openalex.org/C2781395549","wikidata":"https://www.wikidata.org/wiki/Q4680762","display_name":"Adaptive sampling","level":3,"score":0.4426000118255615},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.43779999017715454},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.42739999294281006},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.41990000009536743},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3758000135421753},{"id":"https://openalex.org/C29202148","wikidata":"https://www.wikidata.org/wiki/Q287260","display_name":"Resource allocation","level":2,"score":0.3619000017642975},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.35010001063346863},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32269999384880066},{"id":"https://openalex.org/C93361087","wikidata":"https://www.wikidata.org/wiki/Q4426698","display_name":"Data consistency","level":2,"score":0.3041999936103821},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.30059999227523804},{"id":"https://openalex.org/C27405340","wikidata":"https://www.wikidata.org/wiki/Q440296","display_name":"Image scaling","level":4,"score":0.2800999879837036},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.27399998903274536},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.2667999863624573},{"id":"https://openalex.org/C2989087649","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Image synthesis","level":3,"score":0.2515999972820282}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.00141","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00141","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.2603.00141","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.00141","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Image":[0],"Chain-of-Thought":[1],"(Image-CoT)":[2],"is":[3,29,34],"a":[4,98],"test-time":[5,83],"scaling":[6,84],"paradigm":[7],"that":[8,102,117,138,160],"improves":[9],"image":[10,27,39],"generation":[11],"by":[12,36,134],"extending":[13],"inference":[14],"time.":[15],"Most":[16],"Image-CoT":[17,49],"methods":[18],"focus":[19],"on":[20,107,147],"text-to-image":[21],"(T2I)":[22],"generation.":[23],"Unlike":[24],"T2I":[25],"generation,":[26],"editing":[28,88,150],"goal-directed:":[30],"the":[31,37],"solution":[32],"space":[33],"constrained":[35],"source":[38],"and":[40,66,90,121,128],"instruction.":[41],"This":[42],"mismatch":[43],"causes":[44],"three":[45,94,148,157],"challenges":[46],"when":[47,140],"applying":[48],"to":[50,86,124],"editing:":[51],"inefficient":[52],"resource":[53,100],"allocation":[54,101],"with":[55,174],"fixed":[56],"sampling":[57,168],"budgets,":[58,169],"unreliable":[59],"early-stage":[60],"verification":[61,113],"using":[62],"general":[63],"MLLM":[64],"scores,":[65],"redundant":[67],"edited":[68],"results":[69,142],"from":[70],"large-scale":[71],"sampling.":[72],"To":[73],"address":[74],"this,":[75],"we":[76],"propose":[77],"ADaptive":[78],"Edit-CoT":[79],"(ADE-CoT),":[80],"an":[81,135],"on-demand":[82],"framework":[85],"enhance":[87],"efficiency":[89],"performance.":[91],"It":[92],"incorporates":[93],"key":[95],"strategies:":[96],"(1)":[97],"difficulty-aware":[99],"assigns":[103],"dynamic":[104],"budgets":[105],"based":[106],"estimated":[108],"edit":[109],"difficulty;":[110],"(2)":[111],"edit-specific":[112],"in":[114],"early":[115],"pruning":[116],"uses":[118],"region":[119],"localization":[120],"caption":[122],"consistency":[123],"select":[125],"promising":[126],"candidates;":[127],"(3)":[129],"depth-first":[130],"opportunistic":[131],"stopping,":[132],"guided":[133],"instance-specific":[136],"verifier,":[137],"terminates":[139],"intent-aligned":[141],"are":[143],"found.":[144],"Extensive":[145],"experiments":[146],"SOTA":[149],"models":[151],"(Step1X-Edit,":[152],"BAGEL,":[153],"FLUX.1":[154],"Kontext)":[155],"across":[156],"benchmarks":[158],"show":[159],"ADE-CoT":[161,170],"achieves":[162],"superior":[163],"performance-efficiency":[164],"trade-offs.":[165],"With":[166],"comparable":[167],"obtains":[171],"better":[172],"performance":[173],"more":[175],"than":[176],"2x":[177],"speedup":[178],"over":[179],"Best-of-N.":[180]},"counts_by_year":[],"updated_date":"2026-03-04T07:09:34.246503","created_date":"2026-03-04T00:00:00"}