{"id":"https://openalex.org/W4402455243","doi":"https://doi.org/10.1145/3643788.3648013","title":"BOSS: A dataset to train ML-based systems to repair programs with out-of-bounds write flaws","display_name":"BOSS: A dataset to train ML-based systems to repair programs with out-of-bounds write flaws","publication_year":2024,"publication_date":"2024-04-20","ids":{"openalex":"https://openalex.org/W4402455243","doi":"https://doi.org/10.1145/3643788.3648013"},"language":"en","primary_location":{"id":"doi:10.1145/3643788.3648013","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3643788.3648013","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th ACM/IEEE International Workshop on Automated Program Repair","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/A5107132939","display_name":"Anurag Swarnim Yadav","orcid":"https://orcid.org/0009-0005-7611-2329"},"institutions":[{"id":"https://openalex.org/I33213144","display_name":"University of Florida","ror":"https://ror.org/02y3ad647","country_code":"US","type":"education","lineage":["https://openalex.org/I33213144"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anurag Swarnim Yadav","raw_affiliation_strings":["University of Florida, Gainesville, Florida, USA"],"raw_orcid":"https://orcid.org/0009-0005-7611-2329","affiliations":[{"raw_affiliation_string":"University of Florida, Gainesville, Florida, USA","institution_ids":["https://openalex.org/I33213144"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5102907669","display_name":"Joseph N. Wilson","orcid":"https://orcid.org/0000-0002-5833-8499"},"institutions":[{"id":"https://openalex.org/I33213144","display_name":"University of Florida","ror":"https://ror.org/02y3ad647","country_code":"US","type":"education","lineage":["https://openalex.org/I33213144"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Joseph N. Wilson","raw_affiliation_strings":["University of Florida, Gainesville, Florida, USA"],"raw_orcid":"https://orcid.org/0000-0002-5833-8499","affiliations":[{"raw_affiliation_string":"University of Florida, Gainesville, Florida, USA","institution_ids":["https://openalex.org/I33213144"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I33213144"],"apc_list":null,"apc_paid":null,"fwci":0.4231,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.66659785,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":95,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"26","last_page":"33"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10743","display_name":"Software Testing and Debugging Techniques","score":0.9979000091552734,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T10743","display_name":"Software Testing and Debugging Techniques","score":0.9979000091552734,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/T11005","display_name":"Radiation Effects in Electronics","score":0.9965000152587891,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.9921000003814697,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/boss","display_name":"Boss","score":0.9723542928695679},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6603370904922485},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3476060926914215},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.19963312149047852},{"id":"https://openalex.org/keywords/mechanical-engineering","display_name":"Mechanical engineering","score":0.10329979658126831}],"concepts":[{"id":"https://openalex.org/C2777020290","wikidata":"https://www.wikidata.org/wiki/Q4947493","display_name":"Boss","level":2,"score":0.9723542928695679},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6603370904922485},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3476060926914215},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.19963312149047852},{"id":"https://openalex.org/C78519656","wikidata":"https://www.wikidata.org/wiki/Q101333","display_name":"Mechanical engineering","level":1,"score":0.10329979658126831}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3643788.3648013","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3643788.3648013","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 5th ACM/IEEE International Workshop on Automated Program Repair","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W93013710","https://openalex.org/W103065951","https://openalex.org/W1520957802","https://openalex.org/W1579850852","https://openalex.org/W1976878954","https://openalex.org/W2028820179","https://openalex.org/W2047764386","https://openalex.org/W2350778671","https://openalex.org/W2515236103","https://openalex.org/W2560041978","https://openalex.org/W2613534458","https://openalex.org/W2795192879","https://openalex.org/W2884642766","https://openalex.org/W2912109778","https://openalex.org/W2916006047","https://openalex.org/W2962960733","https://openalex.org/W2964097210","https://openalex.org/W3033053557","https://openalex.org/W3091588759","https://openalex.org/W3133304533","https://openalex.org/W3172196883","https://openalex.org/W3183469243","https://openalex.org/W3205186298","https://openalex.org/W4234020632","https://openalex.org/W4286331360","https://openalex.org/W4308641648","https://openalex.org/W6737694244"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2748952813","https://openalex.org/W3210524396","https://openalex.org/W1996775904","https://openalex.org/W2136244850","https://openalex.org/W1496945489","https://openalex.org/W2092774949","https://openalex.org/W3007787046","https://openalex.org/W2384794170","https://openalex.org/W2376068818"],"abstract_inverted_index":{"C":[0,32],"and":[1,24,33,77,117,137,158,194,271],"C++":[2,34],"are":[3,36,42,269],"widely-used,":[4],"mature":[5],"programming":[6],"languages.":[7],"They":[8],"have":[9,201],"been":[10],"extensively":[11],"used":[12,123],"in":[13,44,62,75,99,192,213,238,247],"development":[14],"of":[15,68,114,128,135,150,169,197,235,243],"projects":[16],"such":[17],"as":[18,41,253],"Linux,":[19],"Windows,":[20],"YouTube,":[21],"Adobe,":[22],"Firefox,":[23],"Google":[25],"Chrome.":[26],"Due":[27],"to":[28,38,109,124,131,260,288],"poor":[29],"memory":[30],"safety,":[31],"programs":[35,43,64,119,152],"vulnerable":[37],"security":[39,227],"attacks":[40],"languages":[45],"that":[46,96,120,203,262],"depend":[47],"on":[48,178],"C/C++":[49,63,151],"library":[50],"code.":[51],"As":[52],"per":[53],"the":[54,66,69,83,133,222,236,263],"Common":[55],"Weakness":[56],"Enumeration":[57],"(CWE),":[58],"out-of-bounds":[59],"(OOB)":[60],"write":[61,81,161],"topped":[65],"list":[67],"25":[70],"most":[71],"dangerous":[72],"software":[73],"weakness":[74],"2021":[76],"2022.":[78],"Fixing":[79],"OOB":[80,139,160,184],"at":[82],"source":[84],"code":[85],"level":[86],"still":[87],"requires":[88],"human":[89],"experts.":[90],"This":[91],"is":[92,274],"a":[93,107,111,148,167,239],"tedious":[94],"task":[95],"may":[97],"result":[98],"erroneous":[100],"programs.":[101,172],"In":[102,173,220],"this":[103,174],"paper":[104,175],"we":[105,176],"propose":[106,284],"technique":[108,143],"create":[110],"data":[112],"set":[113,149,168,218,242],"corresponding":[115,170],"flawed":[116,171],"correct":[118],"can":[121,208],"be":[122],"perform":[125],"supervised":[126],"training":[127,277],"deep-learning":[129],"models":[130],"automate":[132],"process":[134],"detecting":[136],"patching":[138],"writes.":[140],"The":[141],"proposed":[142],"has":[144],"two":[145,285],"elements:":[146],"collecting":[147],"from":[153],"online":[154],"sources":[155],"(correct":[156],"programs)":[157],"injecting":[159],"errors":[162],"into":[163],"them,":[164],"thus":[165],"yielding":[166],"focus":[177],"four":[179],"main":[180],"flaws":[181],"associated":[182,291],"with":[183,292],"writes:":[185],"faulty":[186,188,190,195],"access,":[187],"declaration,":[189],"guard":[191],"loops,":[193],"usage":[196],"memory-write":[198],"APIs.":[199],"We":[200,282],"found":[202],"popular":[204],"fault":[205],"localization":[206],"tools":[207],"not":[209,232],"localize":[210],"complicated":[211],"bugs":[212,237,264],"our":[214,267,272],"buffer":[215],"overflow":[216],"sample":[217],"(BOSS).":[219],"addition,":[221],"current":[223],"state-of-the-art":[224],"machine":[225],"learning":[226],"flaw":[228],"repair":[229,233,280],"tool":[230,268],"could":[231],"any":[234],"randomly":[240],"selected":[241],"BOSS":[244],"samples":[245],"and,":[246],"some":[248],"cases,":[249],"generated":[250],"out-of-bound":[251],"writes":[252],"suggested":[254],"patches.":[255],"These":[256],"results":[257],"lead":[258],"us":[259],"conclude":[261],"injected":[265],"by":[266],"significant":[270],"dataset":[273],"useful":[275],"for":[276],"neural":[278],"program":[279],"models.":[281],"also":[283],"data-augmentation":[286],"techniques":[287],"overcome":[289],"problems":[290],"limited-size":[293],"corpora.":[294]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2025-10-10T00:00:00"}