{"id":"https://openalex.org/W7140656118","doi":"https://doi.org/10.48550/arxiv.2603.24119","title":"Enhancing and Reporting Robustness Boundary of Neural Code Models for Intelligent Code Understanding","display_name":"Enhancing and Reporting Robustness Boundary of Neural Code Models for Intelligent Code Understanding","publication_year":2026,"publication_date":"2026-03-25","ids":{"openalex":"https://openalex.org/W7140656118","doi":"https://doi.org/10.48550/arxiv.2603.24119"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.24119","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.24119","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.24119","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021206073","display_name":"Tingxu Han","orcid":"https://orcid.org/0000-0003-1821-611X"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Han, Tingxu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130712444","display_name":"Wei Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Wei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130709852","display_name":"Weisong Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Weisong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130645785","display_name":"Hao Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Hao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130666656","display_name":"Chunrong Fang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Chunrong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130696977","display_name":"Yuan Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Yuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130646013","display_name":"Xiaofang Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xiaofang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130714819","display_name":"Zhenyu Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Zhenyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130703197","display_name":"Yang Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yang","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"corresponding_author_ids":["https://openalex.org/A5021206073"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.608299970626831,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.608299970626831,"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/T11241","display_name":"Advanced Malware Detection Techniques","score":0.22619999945163727,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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":0.05550000071525574,"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/robustness","display_name":"Robustness (evolution)","score":0.8263999819755554},{"id":"https://openalex.org/keywords/adversarial-system","display_name":"Adversarial system","score":0.5112000107765198},{"id":"https://openalex.org/keywords/smoothing","display_name":"Smoothing","score":0.48750001192092896},{"id":"https://openalex.org/keywords/source-code","display_name":"Source code","score":0.41190001368522644},{"id":"https://openalex.org/keywords/certification","display_name":"Certification","score":0.40459999442100525},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.38589999079704285}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.8263999819755554},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7249000072479248},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.5112000107765198},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.48750001192092896},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43549999594688416},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.41190001368522644},{"id":"https://openalex.org/C46304622","wikidata":"https://www.wikidata.org/wiki/Q374814","display_name":"Certification","level":2,"score":0.40459999442100525},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3919999897480011},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3889000117778778},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.38589999079704285},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.36399999260902405},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.3578000068664551},{"id":"https://openalex.org/C137726913","wikidata":"https://www.wikidata.org/wiki/Q7353550","display_name":"Robustness testing","level":3,"score":0.3133000135421753},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.2906000018119812},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.28450000286102295}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.24119","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.24119","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.24119","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.24119","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.6478132605552673}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"With":[0],"the":[1,87,131,189,223],"development":[2],"of":[3,245],"deep":[4],"learning,":[5],"Neural":[6],"Code":[7],"Models":[8],"(NCMs)":[9],"such":[10,82],"as":[11,83,156],"CodeBERT":[12],"and":[13,25,74,97,109,167],"CodeLlama":[14],"are":[15,35,67,257],"widely":[16],"used":[17],"for":[18,114],"code":[19,26,127],"understanding":[20],"tasks,":[21],"including":[22],"defect":[23,185,219],"detection":[24],"classification.":[27],"However,":[28],"recent":[29],"studies":[30],"have":[31],"revealed":[32],"that":[33,44,171,205,248],"NCMs":[34],"vulnerable":[36],"to":[37,51,61,79,104,125,129,139,194,228,251],"adversarial":[38,99,147,249],"examples,":[39],"inputs":[40],"with":[41,196,230],"subtle":[42],"perturbations":[43,124],"induce":[45,151],"incorrect":[46],"predictions":[47],"while":[48,178,212],"remaining":[49],"difficult":[50],"detect.":[52],"Existing":[53],"defenses":[54],"address":[55,86],"this":[56],"issue":[57],"via":[58],"data":[59],"augmentation":[60],"empirically":[62],"improve":[63],"robustness,":[64],"but":[65],"they":[66],"costly,":[68],"offer":[69],"no":[70,252],"theoretical":[71],"robustness":[72,108,112,143,243],"guarantees,":[73],"typically":[75],"require":[76],"white-box":[77],"access":[78],"model":[80],"internals,":[81],"gradients.":[84],"To":[85],"above":[88],"challenges,":[89],"we":[90],"propose":[91],"ENBECOME,":[92],"a":[93,142,153,198,232],"novel":[94],"black-box":[95],"training-free":[96],"lightweight":[98],"defense.":[100],"ENBECOME":[101,116,138,172,206,238],"is":[102],"designed":[103],"both":[105],"enhance":[106],"empirical":[107],"report":[110],"certified":[111,157,242],"boundaries":[113],"NCMs.":[115],"operates":[117],"solely":[118],"during":[119],"inference,":[120],"introducing":[121],"random,":[122],"semantics-preserving":[123],"input":[126],"snippets":[128],"smooth":[130],"NCM's":[132],"decision":[133],"boundaries.":[134],"This":[135],"smoothing":[136],"enables":[137],"formally":[140],"certify":[141],"radius":[144,244],"within":[145],"which":[146],"examples":[148],"can":[149],"never":[150],"misclassification,":[152],"property":[154],"known":[155],"robustness.":[158],"We":[159],"conduct":[160],"comprehensive":[161],"experiments":[162],"across":[163],"multiple":[164],"NCM":[165],"architectures":[166],"tasks.":[168],"Results":[169,203],"show":[170,204],"significantly":[173,207],"reduces":[174,188,208,222],"attack":[175,209],"success":[176,210],"rates":[177,211],"maintaining":[179,213],"high":[180,214],"accuracy.":[181,202,215,236],"For":[182,216],"example,":[183,217],"in":[184,201,218,235],"detection,":[186,220],"it":[187,221],"average":[190,224,241],"ASR":[191,225],"from":[192,226],"42.43%":[193,227],"9.74%":[195,229],"only":[197,231],"0.29%":[199,233],"drop":[200,234],"Furthermore,":[237],"achieves":[239],"an":[240],"1.63,":[246],"meaning":[247],"modifications":[250],"more":[253],"than":[254],"1.63":[255],"identifiers":[256],"provably":[258],"ineffective.":[259]},"counts_by_year":[],"updated_date":"2026-03-27T06:05:27.210665","created_date":"2026-03-27T00:00:00"}