{"id":"https://openalex.org/W4409047589","doi":"https://doi.org/10.1145/3721146.3721953","title":"Beyond Test-Time Compute Strategies: Advocating Energy-per-Token in LLM Inference","display_name":"Beyond Test-Time Compute Strategies: Advocating Energy-per-Token in LLM Inference","publication_year":2025,"publication_date":"2025-03-30","ids":{"openalex":"https://openalex.org/W4409047589","doi":"https://doi.org/10.1145/3721146.3721953"},"language":"en","primary_location":{"id":"doi:10.1145/3721146.3721953","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721146.3721953","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721953","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721953","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5064105629","display_name":"Pascal Wilhelm","orcid":null},"institutions":[{"id":"https://openalex.org/I4577782","display_name":"Technische Universit\u00e4t Berlin","ror":"https://ror.org/03v4gjf40","country_code":"DE","type":"education","lineage":["https://openalex.org/I4577782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Patrick Wilhelm","raw_affiliation_strings":["BIFOLD, Berlin, Germany","Technische Universit\u00e4t Berlin, Berlin, Germany"],"raw_orcid":"https://orcid.org/0009-0004-8714-5647","affiliations":[{"raw_affiliation_string":"BIFOLD, Berlin, Germany","institution_ids":[]},{"raw_affiliation_string":"Technische Universit\u00e4t Berlin, Berlin, Germany","institution_ids":["https://openalex.org/I4577782"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086364217","display_name":"Thorsten Wittkopp","orcid":"https://orcid.org/0000-0001-5154-7813"},"institutions":[{"id":"https://openalex.org/I4577782","display_name":"Technische Universit\u00e4t Berlin","ror":"https://ror.org/03v4gjf40","country_code":"DE","type":"education","lineage":["https://openalex.org/I4577782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Thorsten Wittkopp","raw_affiliation_strings":["Technische Universit\u00e4t Berlin, Berlin, Germany"],"raw_orcid":"https://orcid.org/0000-0001-5154-7813","affiliations":[{"raw_affiliation_string":"Technische Universit\u00e4t Berlin, Berlin, Germany","institution_ids":["https://openalex.org/I4577782"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5042349846","display_name":"Odej Kao","orcid":"https://orcid.org/0000-0001-6454-6799"},"institutions":[{"id":"https://openalex.org/I4577782","display_name":"Technische Universit\u00e4t Berlin","ror":"https://ror.org/03v4gjf40","country_code":"DE","type":"education","lineage":["https://openalex.org/I4577782"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Odej Kao","raw_affiliation_strings":["Technische Universit\u00e4t Berlin, Berlin, Germany"],"raw_orcid":"https://orcid.org/0000-0001-6454-6799","affiliations":[{"raw_affiliation_string":"Technische Universit\u00e4t Berlin, Berlin, Germany","institution_ids":["https://openalex.org/I4577782"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":18.2286,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.9876339,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"208","last_page":"215"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.9693999886512756,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T11986","display_name":"Scientific Computing and Data Management","score":0.9693999886512756,"subfield":{"id":"https://openalex.org/subfields/1802","display_name":"Information Systems and Management"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.958899974822998,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.9391000270843506,"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/security-token","display_name":"Security token","score":0.7498904466629028},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6688967347145081},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6237822771072388},{"id":"https://openalex.org/keywords/test","display_name":"Test (biology)","score":0.6153275966644287},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.27064791321754456},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.23987185955047607}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7498904466629028},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6688967347145081},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6237822771072388},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.6153275966644287},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.27064791321754456},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.23987185955047607},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3721146.3721953","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721146.3721953","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721953","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th Workshop on Machine Learning and Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3721146.3721953","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3721146.3721953","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3721146.3721953","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th Workshop on Machine Learning and Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4409047589.pdf"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W2963122961","https://openalex.org/W4281971810","https://openalex.org/W4308245305","https://openalex.org/W4366564134","https://openalex.org/W4380322001","https://openalex.org/W4380353763","https://openalex.org/W4387321091","https://openalex.org/W4390189088","https://openalex.org/W4406650295","https://openalex.org/W6852874933","https://openalex.org/W6892464000"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W4388335561","https://openalex.org/W2970530566","https://openalex.org/W4288261899","https://openalex.org/W4307309205","https://openalex.org/W2967478618","https://openalex.org/W4385009901","https://openalex.org/W4385572700"],"abstract_inverted_index":{"Large":[0],"Language":[1,38],"Models":[2,39],"(LLMs)":[3],"demonstrate":[4],"exceptional":[5],"performance":[6,68],"across":[7],"diverse":[8],"tasks":[9],"but":[10],"come":[11],"with":[12,52,134],"substantial":[13],"energy":[14,84,125,140],"and":[15,29,181],"computational":[16,75],"costs,":[17,85],"particularly":[18],"in":[19,95,122,158],"request-heavy":[20],"scenarios.":[21],"In":[22],"many":[23],"real-world":[24],"applications,":[25],"the":[26,67,107,113],"full":[27],"scale":[28],"capabilities":[30],"of":[31,69,117],"LLMs":[32],"are":[33],"often":[34],"unnecessary,":[35],"as":[36,57,145],"Small":[37],"(SLMs)":[40],"can":[41,65,80],"provide":[42],"accurate":[43],"responses":[44],"for":[45,99,128,186],"simpler":[46],"text":[47],"generation":[48],"tasks.":[49],"When":[50],"enhanced":[51],"advanced":[53],"reasoning":[54,157,167],"strategies,":[55],"such":[56],"Chain-of-Thought":[58],"(CoT)":[59],"prompting":[60],"or":[61],"Majority":[62],"Voting,":[63],"SLMs":[64],"approach":[66],"larger":[70,104],"models":[71,101],"while":[72],"reducing":[73],"overall":[74],"requirements.":[76],"However,":[77],"these":[78,93],"strategies":[79,98,183],"also":[81],"introduce":[82],"additional":[83],"creating":[86],"an":[87],"energy-accuracy":[88],"trade-off.":[89],"Our":[90],"analysis":[91],"examines":[92],"trade-offs":[94],"test-time":[96],"compute":[97],"smaller":[100],"compared":[102],"to":[103,147,165],"ones,":[105],"using":[106,162],"MMLU":[108],"benchmark.":[109],"Additionally,":[110],"we":[111,138,154],"explore":[112],"input-output":[114],"token":[115,160],"dynamics":[116],"transformer":[118],"architectures,":[119],"which":[120],"result":[121],"nonlinear":[123],"hardware":[124],"operation":[126],"curves":[127,164],"LLMs.":[129],"To":[130],"bridge":[131],"AI":[132,188],"research":[133],"its":[135],"physical":[136],"impact,":[137],"propose":[139,155],"efficiency":[141],"metrics,":[142],"including":[143],"Energy-per-Token,":[144],"complements":[146],"traditional":[148],"accuracy":[149,185],"benchmarks.":[150],"Beyond":[151],"model":[152,179],"selection,":[153],"controlled":[156],"CoT":[159],"generation,":[161],"operating":[163],"regulate":[166],"depth":[168],"dynamically.":[169],"This":[170],"vision":[171],"integrates":[172],"a":[173],"energy-aware":[174],"routing":[175],"mechanism,":[176],"ensuring":[177],"that":[178],"selection":[180],"inference":[182],"balance":[184],"sustainable":[187],"deployment.":[189]},"counts_by_year":[{"year":2026,"cited_by_count":6}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}