{"id":"https://openalex.org/W7163903970","doi":"https://doi.org/10.48550/arxiv.2606.06539","title":"Synthetic Benchmarks Overstate Forward-Forward Scaling: Real-Data Limits of Layer-Local Training","display_name":"Synthetic Benchmarks Overstate Forward-Forward Scaling: Real-Data Limits of Layer-Local Training","publication_year":2026,"publication_date":"2026-06-04","ids":{"openalex":"https://openalex.org/W7163903970","doi":"https://doi.org/10.48550/arxiv.2606.06539"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.06539","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06539","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.06539","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138168063","display_name":"Yucheng Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Chen, Yucheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5138168063"],"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/T10036","display_name":"Advanced Neural Network Applications","score":0.6000999808311462,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.6000999808311462,"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/T11659","display_name":"Advanced Image Fusion Techniques","score":0.07599999755620956,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.06729999929666519,"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/baseline","display_name":"Baseline (sea)","score":0.6582000255584717},{"id":"https://openalex.org/keywords/curse-of-dimensionality","display_name":"Curse of dimensionality","score":0.5435000061988831},{"id":"https://openalex.org/keywords/synthetic-data","display_name":"Synthetic data","score":0.41749998927116394},{"id":"https://openalex.org/keywords/dimension","display_name":"Dimension (graph theory)","score":0.4163999855518341},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4133000075817108},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.4050999879837036},{"id":"https://openalex.org/keywords/ceiling","display_name":"Ceiling (cloud)","score":0.3970000147819519},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.388700008392334}],"concepts":[{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.6582000255584717},{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.5435000061988831},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5286999940872192},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4885999858379364},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.41749998927116394},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.4163999855518341},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4133000075817108},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.4050999879837036},{"id":"https://openalex.org/C2777489069","wikidata":"https://www.wikidata.org/wiki/Q1589822","display_name":"Ceiling (cloud)","level":2,"score":0.3970000147819519},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3919000029563904},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.388700008392334},{"id":"https://openalex.org/C199521495","wikidata":"https://www.wikidata.org/wiki/Q181487","display_name":"Audit","level":2,"score":0.38580000400543213},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.37040001153945923},{"id":"https://openalex.org/C139676723","wikidata":"https://www.wikidata.org/wiki/Q1193832","display_name":"Sign (mathematics)","level":2,"score":0.36410000920295715},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.36250001192092896},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.3433000147342682},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.311599999666214},{"id":"https://openalex.org/C2779439359","wikidata":"https://www.wikidata.org/wiki/Q317088","display_name":"Commodity","level":2,"score":0.2980000078678131},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.29319998621940613},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.29249998927116394},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.2822999954223633},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.2797999978065491},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.27880001068115234},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.26330000162124634},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26190000772476196},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2508000135421753},{"id":"https://openalex.org/C13736549","wikidata":"https://www.wikidata.org/wiki/Q4489420","display_name":"Industrial engineering","level":1,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.06539","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06539","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.06539","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06539","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":"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"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.4006802439689636,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Forward-Forward":[0],"(FF)":[1],"learning":[2],"[Hinton,":[3],"2022]":[4],"replaces":[5],"backpropagation":[6],"with":[7,114,179,198],"strictly":[8],"layer-local":[9,29,87],"goodness":[10],"updates.":[11],"Recent":[12],"FF-CNN":[13],"work":[14],"has":[15],"narrowed":[16],"the":[17,25,64,73,111,119,126,172],"gap":[18,112,174],"to":[19,83],"BP":[20,135,158],"on":[21,108,164,169,219],"32x32":[22],"benchmarks,":[23],"raising":[24],"question":[26],"of":[27,63],"whether":[28],"training":[30,88],"is":[31,248],"becoming":[32],"a":[33,144,240],"viable":[34],"alternative":[35],"at":[36,77,130,148,245],"realistic":[37],"scale.":[38],"To":[39],"probe":[40,187],"this":[41,131,246],"rigorously,":[42],"we":[43],"develop":[44],"DTG-FF":[45,104,155],"--":[46,55,125,142],"dynamic":[47],"temperature":[48],"goodness,":[49],"decoupled":[50],"normalization,":[51],"and":[52,72,80,97,110,177,202],"multi-layer":[53],"fusion":[54],"as":[56,159],"an":[57,99],"instrument":[58,121],"that":[59],"sets":[60],"FF-family":[61],"state":[62],"art":[65],"across":[66],"nine":[67],"real-data":[68,145],"benchmarks":[69],"(91.8%":[70],"CIFAR-10":[71],"first":[74,127],"FF":[75,128,205,210,244],"baseline":[76,102,129],"ImageNet-100":[78],"224x224),":[79],"use":[81],"it":[82],"audit":[84],"how":[85],"far":[86],"actually":[89],"scales.":[90],"(1)":[91],"Real-data":[92],"scaling.":[93],"Under":[94],"identical":[95],"recipe":[96],"backbone,":[98],"architecture-matched":[100],"BP-DeepSup":[101],"beats":[103],"by":[105],"2.40/5.93":[106],"pp":[107],"CIFAR-10/CIFAR-100,":[109],"widens":[113,178],"class":[115,160],"count.":[116],"At":[117],"224x224":[118],"same":[120],"reaches":[122,226],"only":[123],"49.4%":[124],"scale,":[132],"versus":[133,232],"typical":[134],"above":[136],"75%":[137],"[Tian":[138],"et":[139],"al.,":[140],"2020]":[141],"exposing":[143],"ceiling":[146],"invisible":[147],"32x32.":[149],"(2)":[150],"Synthetic":[151],"vs.":[152,185],"real":[153,170],"K-conflict.":[154],"increasingly":[156],"outperforms":[157],"count":[161],"K":[162],"grows":[163],"synthetic":[165,193],"teacher-student":[166],"tasks,":[167],"yet":[168],"images":[171],"FF-BP":[173],"reverses":[175],"sign":[176],"K.":[180],"A":[181],"within-dataset":[182],"CIFAR-100":[183],"coarse":[184],"fine":[186],"isolates":[188],"label-hierarchy":[189],"from":[190],"image":[191],"distribution:":[192],"K-sweeps":[194],"confound":[195],"output":[196],"dimensionality":[197],"fine-grained":[199],"discrimination":[200],"difficulty":[201],"thereby":[203],"overstate":[204],"transferability.":[206],"(3)":[207],"Systems":[208],"audit.":[209],"can":[211],"be":[212],"implemented":[213],"without":[214],"storing":[215],"depth-wide":[216],"activations,":[217],"but":[218],"commodity":[220],"8":[221],"GB":[222,228,235],"hardware":[223],"standard":[224],"BP+gradient-accumulation":[225],"4.18":[227],"/":[229,236],"157":[230],"imgs/s":[231],"DTG-FF's":[233],"7.90":[234],"138":[237],"imgs/s,":[238],"so":[239],"memory-based":[241],"justification":[242],"for":[243],"scale":[247],"not":[249],"supported":[250],"under":[251],"fair":[252],"baselines.":[253]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-09T00:00:00"}