{"id":"https://openalex.org/W4406892132","doi":"https://doi.org/10.1109/fllm63129.2024.10852507","title":"Can Separators Improve Chain-of-Thought Prompting?","display_name":"Can Separators Improve Chain-of-Thought Prompting?","publication_year":2024,"publication_date":"2024-11-26","ids":{"openalex":"https://openalex.org/W4406892132","doi":"https://doi.org/10.1109/fllm63129.2024.10852507"},"language":"en","primary_location":{"id":"doi:10.1109/fllm63129.2024.10852507","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fllm63129.2024.10852507","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 2nd International Conference on Foundation and Large Language Models (FLLM)","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/A5083853459","display_name":"Yoonjeong Park","orcid":null},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Yoonjeong Park","raw_affiliation_strings":["Yonsei University,Seoul,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Yonsei University,Seoul,Republic of Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100359572","display_name":"Hyunjin Kim","orcid":"https://orcid.org/0000-0003-2461-9935"},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyunjin Kim","raw_affiliation_strings":["Yonsei University,Seoul,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Yonsei University,Seoul,Republic of Korea","institution_ids":["https://openalex.org/I193775966"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5093957968","display_name":"Chanyeol Choi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chanyeol Choi","raw_affiliation_strings":["Linq,Cambridge,Massachusetts,United States"],"affiliations":[{"raw_affiliation_string":"Linq,Cambridge,Massachusetts,United States","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019877980","display_name":"Junseong Kim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Junseong Kim","raw_affiliation_strings":["Linq,Cambridge,Massachusetts,United States"],"affiliations":[{"raw_affiliation_string":"Linq,Cambridge,Massachusetts,United States","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5114087128","display_name":"J. Sohn","orcid":null},"institutions":[{"id":"https://openalex.org/I193775966","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96","country_code":"KR","type":"education","lineage":["https://openalex.org/I193775966"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jy-Yong Sohn","raw_affiliation_strings":["Yonsei University,Seoul,Republic of Korea"],"affiliations":[{"raw_affiliation_string":"Yonsei University,Seoul,Republic of Korea","institution_ids":["https://openalex.org/I193775966"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5083853459"],"corresponding_institution_ids":["https://openalex.org/I193775966"],"apc_list":null,"apc_paid":null,"fwci":0.3637,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.7104313,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"493","last_page":"500"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.3183000087738037,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.3183000087738037,"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/chain","display_name":"Chain (unit)","score":0.5324071049690247},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5316897630691528}],"concepts":[{"id":"https://openalex.org/C199185054","wikidata":"https://www.wikidata.org/wiki/Q552299","display_name":"Chain (unit)","level":2,"score":0.5324071049690247},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5316897630691528},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C1276947","wikidata":"https://www.wikidata.org/wiki/Q333","display_name":"Astronomy","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/fllm63129.2024.10852507","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fllm63129.2024.10852507","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 2nd International Conference on Foundation and Large Language Models (FLLM)","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":0,"referenced_works":[],"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/W4396696052"],"abstract_inverted_index":{"Chain-of-thought":[0],"(CoT)":[1],"prompting":[2],"is":[3,24,163],"a":[4,63],"simple":[5],"and":[6,130,142],"effective":[7],"method":[8,64],"for":[9],"improving":[10],"the":[11,38,42,51,70,84,101,114,125,128,131,146,156],"reasoning":[12,106,157],"capabilities":[13],"of":[14,22,47,54,72,127,133,148,159],"Large":[15],"Language":[16],"Models":[17],"(LLMs).":[18],"The":[19,161],"basic":[20],"idea":[21],"CoT":[23,48,76],"to":[25,82,154],"let":[26],"LLMs":[27,85],"break":[28],"down":[29],"their":[30,87],"thought":[31,88],"processes":[32,89],"step-by-step":[33],"by":[34,57],"putting":[35],"exemplars":[36,46],"in":[37,75],"input":[39],"prompt.":[40],"However,":[41],"densely":[43],"structured":[44],"prompt":[45],"may":[49],"cause":[50],"cognitive":[52],"overload":[53],"LLMs.":[55],"Inspired":[56],"human":[58],"cognition,":[59],"we":[60],"introduce":[61],"COT-SEP,":[62],"that":[65,97],"strategically":[66],"employs":[67],"separators":[68,79,134,149],"at":[69,165],"end":[71],"each":[73],"exemplar":[74],"prompting.":[77],"These":[78],"are":[80],"designed":[81],"help":[83],"understand":[86],"better":[90],"while":[91],"reasoning.":[92],"Interestingly,":[93,145],"it":[94],"turns":[95],"out":[96],"COT-SEP":[98],"significantly":[99],"improves":[100],"LLMs\u2019":[102],"performances":[103],"on":[104,136],"complex":[105],"tasks":[107],"(e.g.,":[108],"GSM8K,":[109],"AQuA,":[110],"CSQA),":[111],"compared":[112],"with":[113],"vanilla":[115],"CoT,":[116],"which":[117],"does":[118],"not":[119],"use":[120],"separators.":[121],"We":[122],"also":[123],"study":[124],"effects":[126],"type":[129],"location":[132],"tested":[135],"multiple":[137],"LLMs,":[138],"including":[139],"GPT-3.5-Turbo,":[140],"GPT-4,":[141],"LLaMA-2":[143],"7B.":[144],"type/location":[147],"should":[150],"be":[151],"chosen":[152],"appropriately":[153],"boost":[155],"capability":[158],"CoT.":[160],"code":[162],"available":[164],"https://github.com/cottonlove/CoT-SEP":[166]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2025-12-27T23:08:20.325037","created_date":"2025-10-10T00:00:00"}