{"id":"https://openalex.org/W4407953060","doi":"https://doi.org/10.1145/3701551.3703563","title":"Improving FIM Code Completions via Context & Curriculum Based Learning","display_name":"Improving FIM Code Completions via Context & Curriculum Based Learning","publication_year":2025,"publication_date":"2025-02-26","ids":{"openalex":"https://openalex.org/W4407953060","doi":"https://doi.org/10.1145/3701551.3703563"},"language":"en","primary_location":{"id":"doi:10.1145/3701551.3703563","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3701551.3703563","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Eighteenth ACM International Conference on Web Search and Data Mining","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/A5092492046","display_name":"Hitesh Sagtani","orcid":"https://orcid.org/0009-0003-6995-1912"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Hitesh Sagtani","raw_affiliation_strings":["Sourcegraph Inc, Bangalore, India"],"affiliations":[{"raw_affiliation_string":"Sourcegraph Inc, Bangalore, India","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018503243","display_name":"Rishabh Mehrotra","orcid":"https://orcid.org/0000-0002-0836-4605"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rishabh Mehrotra","raw_affiliation_strings":["Pavo AI, London, United Kingdom"],"affiliations":[{"raw_affiliation_string":"Pavo AI, London, United Kingdom","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037983013","display_name":"Beyang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Beyang Liu","raw_affiliation_strings":["Sourcegraph Inc, San Francisco, CA, USA"],"affiliations":[{"raw_affiliation_string":"Sourcegraph Inc, San Francisco, CA, USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5092492046"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.03431331,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"801","last_page":"810"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12171","display_name":"Open Education and E-Learning","score":0.9341999888420105,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T12171","display_name":"Open Education and E-Learning","score":0.9341999888420105,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"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/T10260","display_name":"Software Engineering Research","score":0.9078999757766724,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.6274638175964355},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6239156126976013},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.5512616634368896},{"id":"https://openalex.org/keywords/curriculum","display_name":"Curriculum","score":0.5299941897392273},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.421072393655777},{"id":"https://openalex.org/keywords/data-science","display_name":"Data science","score":0.3431551456451416},{"id":"https://openalex.org/keywords/software-engineering","display_name":"Software engineering","score":0.34211763739585876},{"id":"https://openalex.org/keywords/sociology","display_name":"Sociology","score":0.20267254114151},{"id":"https://openalex.org/keywords/pedagogy","display_name":"Pedagogy","score":0.12242370843887329},{"id":"https://openalex.org/keywords/biology","display_name":"Biology","score":0.07766148447990417}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6274638175964355},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6239156126976013},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5512616634368896},{"id":"https://openalex.org/C47177190","wikidata":"https://www.wikidata.org/wiki/Q207137","display_name":"Curriculum","level":2,"score":0.5299941897392273},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.421072393655777},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3431551456451416},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.34211763739585876},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.20267254114151},{"id":"https://openalex.org/C19417346","wikidata":"https://www.wikidata.org/wiki/Q7922","display_name":"Pedagogy","level":1,"score":0.12242370843887329},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.07766148447990417},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"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:10.1145/3701551.3703563","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3701551.3703563","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Eighteenth ACM International Conference on Web Search and Data Mining","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.5799999833106995,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":10,"referenced_works":["https://openalex.org/W2066334462","https://openalex.org/W4319006533","https://openalex.org/W4366204357","https://openalex.org/W4384154386","https://openalex.org/W4389286060","https://openalex.org/W4389955489","https://openalex.org/W4391558462","https://openalex.org/W4400582900","https://openalex.org/W4402714526","https://openalex.org/W6604372272"],"related_works":["https://openalex.org/W2348562106","https://openalex.org/W2370820329","https://openalex.org/W2370554813","https://openalex.org/W2387560707","https://openalex.org/W2363525455","https://openalex.org/W4312355418","https://openalex.org/W4362576712","https://openalex.org/W2314810092","https://openalex.org/W2384329035","https://openalex.org/W2373380871"],"abstract_inverted_index":{"Fill-in-the-Middle":[0],"(FIM)":[1],"models":[2,75,157,169,199,222],"play":[3],"a":[4,84,136,224],"vital":[5],"role":[6],"in":[7,57,206,243],"code":[8,32,49,92,194],"completion":[9,33,50,65,181,195],"tasks,":[10],"leveraging":[11],"both":[12],"prefix":[13],"and":[14,21,54,94,100,114,135,170,202,219,229,248],"suffix":[15],"context":[16,53,96],"to":[17,29,214],"provide":[18],"more":[19,68,164],"accurate":[20],"contextually":[22],"relevant":[23],"suggestions.":[24],"This":[25,183],"paper":[26],"presents":[27],"approaches":[28],"improve":[30],"FIM":[31,48,129],"while":[34,154],"addressing":[35],"the":[36,58,126,131,160,190],"challenge":[37],"of":[38,177,193],"maintaining":[39],"low":[40],"latency":[41,191,255],"for":[42,166],"real-time":[43],"coding":[44],"assistance.":[45],"We":[46,61,107],"enhance":[47],"by":[51,87],"incorporating":[52,172],"curriculum":[55,85,178],"examples":[56,97,174],"training":[59],"process.":[60],"identify":[62],"patterns":[63,90],"where":[64],"suggestions":[66],"fail":[67],"frequently,":[69],"revealing":[70],"complexities":[71],"that":[72,153,171],"smaller":[73,167],"language":[74],"struggle":[76],"with.":[77],"To":[78],"address":[79],"these":[80],"challenges,":[81],"we":[82,232],"develop":[83],"dataset":[86],"extracting":[88],"hard-to-complete":[89],"from":[91,143],"repositories":[93],"generate":[95],"using":[98],"semantic":[99],"static":[101],"analysis":[102],"tools":[103],"(e.g.":[104],"TSC":[105],"compiler).":[106],"fine-tune":[108],"various":[109],"sized":[110],"models,":[111],"including":[112],"StarCoder":[113],"DeepSeek,":[115],"on":[116],"this":[117],"enhanced":[118],"dataset.":[119],"Our":[120],"evaluation":[121,140],"encompasses":[122],"three":[123],"key":[124],"dimensions:":[125],"Santa":[127],"Coder":[128],"task,":[130],"Amazon":[132],"CCEval":[133],"benchmark,":[134],"new":[137],"Multi-Line":[138],"Infilling":[139],"benchmark":[141],"derived":[142],"SWE-bench.":[144],"Comprehensive":[145],"ablation":[146],"studies":[147],"across":[148],"multiple":[149],"model":[150],"sizes":[151],"reveal":[152],"all":[155],"fine-tuned":[156,221],"show":[158],"improvements,":[159],"performance":[161],"gains":[162],"are":[163,211],"pronounced":[165],"parameter":[168],"difficult-to-complete":[173],"as":[175],"part":[176],"learning":[179],"improves":[180],"performance.":[182],"finding":[184],"is":[185],"particularly":[186],"sig-":[187],"nificant":[188],"given":[189,216],"constraints":[192],"tasks.":[196],"While":[197],"larger":[198],"like":[200],"GPT":[201],"Claude":[203],"perform":[204],"well":[205],"multi-":[207],"line":[208],"completions":[209],"but":[210],"prohibitively":[212],"challenging":[213],"use":[215],"high":[217],"latency,":[218],"our":[220,234],"achieve":[223],"balance":[225],"between":[226],"per-":[227],"formance":[228],"latency.":[230],"Finally,":[231],"validate":[233],"approach":[235],"through":[236],"online":[237],"A/B":[238],"testing,":[239],"demonstrating":[240],"tangible":[241],"improvements":[242],"Completion":[244,249],"Acceptance":[245],"Rate":[246,251],"(CAR)":[247],"Persistence":[250],"(CPR),":[252],"with":[253],"zero":[254],"impact.":[256]},"counts_by_year":[],"updated_date":"2025-12-22T23:10:17.713674","created_date":"2025-10-10T00:00:00"}