{"id":"https://openalex.org/W4406612248","doi":"https://doi.org/10.1109/smc54092.2024.10831379","title":"A Semantic Verifier for Optimizing Small-Scale Large Language Models on Reasoning Tasks","display_name":"A Semantic Verifier for Optimizing Small-Scale Large Language Models on Reasoning Tasks","publication_year":2024,"publication_date":"2024-10-06","ids":{"openalex":"https://openalex.org/W4406612248","doi":"https://doi.org/10.1109/smc54092.2024.10831379"},"language":"en","primary_location":{"id":"doi:10.1109/smc54092.2024.10831379","is_oa":false,"landing_page_url":"https://doi.org/10.1109/smc54092.2024.10831379","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101669635","display_name":"Yu Bai","orcid":"https://orcid.org/0000-0002-4100-8223"},"institutions":[{"id":"https://openalex.org/I125904092","display_name":"Shenyang Aerospace University","ror":"https://ror.org/02423gm04","country_code":"CN","type":"education","lineage":["https://openalex.org/I125904092"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu Bai","raw_affiliation_strings":["Shenyang Aerospace University,School of Computer Science,Shenyang,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shenyang Aerospace University,School of Computer Science,Shenyang,China","institution_ids":["https://openalex.org/I125904092"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100362041","display_name":"Jun Li","orcid":"https://orcid.org/0000-0003-1613-9448"},"institutions":[{"id":"https://openalex.org/I125904092","display_name":"Shenyang Aerospace University","ror":"https://ror.org/02423gm04","country_code":"CN","type":"education","lineage":["https://openalex.org/I125904092"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jun Li","raw_affiliation_strings":["Shenyang Aerospace University,School of Computer Science,Shenyang,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shenyang Aerospace University,School of Computer Science,Shenyang,China","institution_ids":["https://openalex.org/I125904092"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067279159","display_name":"Fang Cai","orcid":null},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fang Cai","raw_affiliation_strings":["Stanford University,School of Humanities and Sciences,Stanford,USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University,School of Humanities and Sciences,Stanford,USA","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101625765","display_name":"Yiyu Liu","orcid":"https://orcid.org/0009-0007-7811-3290"},"institutions":[{"id":"https://openalex.org/I125904092","display_name":"Shenyang Aerospace University","ror":"https://ror.org/02423gm04","country_code":"CN","type":"education","lineage":["https://openalex.org/I125904092"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuting Liu","raw_affiliation_strings":["Shenyang Aerospace University,School of Computer Science,Shenyang,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shenyang Aerospace University,School of Computer Science,Shenyang,China","institution_ids":["https://openalex.org/I125904092"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.3055,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.68539903,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":91,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"2200","last_page":"2205"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.7670999765396118,"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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.7670999765396118,"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/T10028","display_name":"Topic Modeling","score":0.7211999893188477,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.6416000127792358,"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/computer-science","display_name":"Computer science","score":0.8401503562927246},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.5560010075569153},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.543113648891449},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4324357509613037},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.41133278608322144}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8401503562927246},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.5560010075569153},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.543113648891449},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4324357509613037},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.41133278608322144},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/smc54092.2024.10831379","is_oa":false,"landing_page_url":"https://doi.org/10.1109/smc54092.2024.10831379","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":16,"referenced_works":["https://openalex.org/W2251935656","https://openalex.org/W3156636935","https://openalex.org/W3170403598","https://openalex.org/W4366596238","https://openalex.org/W4385569968","https://openalex.org/W4385572867","https://openalex.org/W4386566508","https://openalex.org/W4389043118","https://openalex.org/W6755829550","https://openalex.org/W6778883912","https://openalex.org/W6779068807","https://openalex.org/W6809646742","https://openalex.org/W6811284106","https://openalex.org/W6837989031","https://openalex.org/W6838865847","https://openalex.org/W6845413636"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W3204019825"],"abstract_inverted_index":{"Large":[0,46],"language":[1,17],"models":[2],"(LLMs)":[3],"with":[4,55],"more":[5],"than":[6,57,75],"100":[7],"billion":[8,59],"parameters":[9],"have":[10,20],"revolutionized":[11],"various":[12],"tasks":[13,116,156],"related":[14],"to":[15,67,143],"natural":[16],"processing":[18],"and":[19,131,168],"had":[21],"a":[22,95,118,133,138],"profound":[23],"impact":[24],"in":[25,34,41,77,108,114],"the":[26,35,86,90,110,163],"field":[27],"of":[28,88,92,98,112,166],"artificial":[29],"intelligence.":[30],"However,":[31,69],"deploying":[32],"LLMs":[33,54,76,113],"real":[36],"world":[37],"could":[38,61],"also":[39],"result":[40],"increased":[42],"production":[43,64],"costs.":[44],"Small-scale":[45],"Language":[47],"Models":[48],"(SLLMs),":[49],"which":[50],"are":[51],"smaller,":[52],"compact":[53],"fewer":[56],"10":[58],"parameters,":[60],"significantly":[62],"reduce":[63],"costs":[65],"compared":[66],"LLMs.":[68],"they":[70],"typically":[71],"perform":[72],"less":[73],"effectively":[74,161],"general.":[78],"Although":[79],"In-context":[80],"learning":[81],"prompting":[82],"has":[83],"successfully":[84],"enhanced":[85],"capabilities":[87],"SLLMs,":[89,167],"construction":[91],"prompts":[93,123],"requires":[94],"certain":[96],"level":[97],"human":[99],"expertise.":[100],"In":[101],"this":[102],"study,":[103],"we":[104],"explore":[105],"enhancing":[106],"SLLMs":[107,130],"emulating":[109],"performance":[111,165],"reasoning":[115,145,155],"at":[117],"minimal":[119],"cost,":[120],"without":[121],"any":[122],"provided":[124],"by":[125],"humans.":[126],"We":[127],"employ":[128],"two":[129],"incorporate":[132],"ranking":[134],"model":[135],"based":[136],"on":[137,149],"Semantic":[139],"Verifier":[140],"between":[141],"them":[142],"facilitate":[144],"tasks.":[146],"Experiments":[147],"conducted":[148],"four":[150],"publicly":[151],"available":[152],"datasets":[153],"for":[154],"demonstrate":[157],"that":[158],"our":[159],"approach":[160],"enhances":[162],"inference":[164],"it":[169],"achieves":[170],"new":[171],"state-of-the-art":[172],"results.":[173]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}