{"id":"https://openalex.org/W7166867806","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1766","title":"FLOP-Efficient Training: Early Stopping Based on Test-Time Compute Awareness","display_name":"FLOP-Efficient Training: Early Stopping Based on Test-Time Compute Awareness","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166867806","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1766"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1766","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1766","pdf_url":"https://aclanthology.org/2026.findings-acl.1766.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1766.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5022851730","display_name":"Hossam Amer","orcid":"https://orcid.org/0000-0003-1136-227X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hossam Amer","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139820163","display_name":"Maryam Dialameh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maryam Dialameh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003110976","display_name":"Hossein Rajabzadeh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hossein Rajabzadeh","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139769700","display_name":"Walid Ahmed","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Walid Ahmed","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139728257","display_name":"Weiwei Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Weiwei Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139815085","display_name":"Yang Janet Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang Liu","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.87855373,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"35431","last_page":"35449"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.07609999924898148,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.07609999924898148,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.05920000001788139,"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"}},{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.05640000104904175,"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/feature","display_name":"Feature (linguistics)","score":0.2493000030517578},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.24230000376701355},{"id":"https://openalex.org/keywords/event","display_name":"Event (particle physics)","score":0.2345999926328659},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.2345999926328659},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.23250000178813934}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5629000067710876},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36399999260902405},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.28040000796318054},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2493000030517578},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.24230000376701355},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.23510000109672546},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.2345999926328659},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.2345999926328659},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.23250000178813934},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.22579999268054962}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1766","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1766","pdf_url":"https://aclanthology.org/2026.findings-acl.1766.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1766","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1766","pdf_url":"https://aclanthology.org/2026.findings-acl.1766.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166867806.pdf","grobid_xml":"https://content.openalex.org/works/W7166867806.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Scaling":[0],"training":[1,19,75,96,130,138,154],"compute,":[2],"measured":[3],"in":[4,137,158],"FLOPs,":[5],"has":[6],"long":[7],"been":[8],"shown":[9],"to":[10,36,94,134],"improve":[11],"the":[12,64],"accuracy":[13,65],"of":[14,66],"large":[15],"language":[16],"models,":[17],"yet":[18],"remains":[20],"resource-intensive.Prior":[21],"work":[22],"shows":[23],"that":[24,86,111,121],"increasing":[25],"test-time":[26],"compute":[27,97,126,157],"(TTC)-for":[28],"example":[29],"through":[30],"iterative":[31],"sampling-can":[32],"allow":[33],"smaller":[34],"models":[35],"rival":[37],"or":[38,62],"surpass":[39],"much":[40],"larger":[41],"ones":[42],"at":[43],"lower":[44],"overall":[45],"cost.We":[46],"introduce":[47],"TTCaware":[48],"training,":[49],"where":[50],"an":[51,82,106],"intermediate":[52],"checkpoint":[53,90],"and":[54,91,115,142,155,165],"a":[55,67,89,118,149],"corresponding":[56],"TTC":[57,92,108],"configuration":[58,93],"can":[59],"together":[60],"match":[61],"exceed":[63],"fully":[68],"trained":[69],"model":[70,159,168],"while":[71,140],"requiring":[72],"substantially":[73],"fewer":[74],"FLOPs.Building":[76],"on":[77],"this":[78,102],"insight,":[79],"we":[80,104,116],"propose":[81],"early":[83],"stopping":[84],"algorithm":[85],"jointly":[87],"selects":[88],"minimize":[95],"without":[98],"sacrificing":[99],"accuracy.To":[100],"make":[101],"practical,":[103],"develop":[105],"efficient":[107],"evaluation":[109],"method":[110],"avoids":[112],"exhaustive":[113],"search,":[114],"formalize":[117],"break-even":[119],"bound":[120],"identifies":[122],"when":[123],"increased":[124],"inference":[125,156],"compensates":[127],"for":[128,152],"reduced":[129],"compute.Experiments":[131],"demonstrate":[132],"up":[133],"92%":[135],"reductions":[136],"FLOPs":[139],"maintaining":[141],"sometimes":[143],"remarkably":[144],"improving":[145],"accuracy.These":[146],"results":[147],"highlight":[148],"new":[150],"perspective":[151],"balancing":[153],"development,":[160],"enabling":[161],"faster":[162],"deployment":[163],"cycles":[164],"more":[166],"frequent":[167],"refreshes.":[169]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}