Towards More Trustworthy Deep Code Models by Enabling Out-of-Distribution Detection

Numerous machine learning (ML) models have been developed, including those for software engineering (SE) tasks, under the assumption that training and testing data come from the same distribution. However, training and testing distributions often differ, as training datasets rarely encompass the ent...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings / International Conference on Software Engineering pp. 769 - 781
Main Authors: Yan, Yanfu, Duong, Viet, Shao, Huajie, Poshyvanyk, Denys
Format: Conference Proceeding
Language:English
Published: IEEE 26.04.2025
Subjects:
ISSN:1558-1225
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Numerous machine learning (ML) models have been developed, including those for software engineering (SE) tasks, under the assumption that training and testing data come from the same distribution. However, training and testing distributions often differ, as training datasets rarely encompass the entire distribution, while testing distribution tends to shift over time. Hence, when confronted with out-of-distribution (OOD) instances that differ from the training data, a reliable and trustworthy SE ML model must be capable of detecting them to either abstain from making predictions, or potentially forward these OODs to appropriate models handling other categories or tasks. In this paper, we develop two types of SE-specific OOD detection models, unsupervised and weakly-supervised OOD detection for code. The unsupervised OOD detection approach is trained solely on in-distribution samples while the weakly-supervised approach utilizes a tiny number of OOD samples to further enhance the detection performance in various OOD scenarios. Extensive experimental results demonstrate that our proposed methods significantly outperform the baselines in detecting OOD samples from four different scenarios simultaneously and also positively impact a main code understanding task.
AbstractList Numerous machine learning (ML) models have been developed, including those for software engineering (SE) tasks, under the assumption that training and testing data come from the same distribution. However, training and testing distributions often differ, as training datasets rarely encompass the entire distribution, while testing distribution tends to shift over time. Hence, when confronted with out-of-distribution (OOD) instances that differ from the training data, a reliable and trustworthy SE ML model must be capable of detecting them to either abstain from making predictions, or potentially forward these OODs to appropriate models handling other categories or tasks. In this paper, we develop two types of SE-specific OOD detection models, unsupervised and weakly-supervised OOD detection for code. The unsupervised OOD detection approach is trained solely on in-distribution samples while the weakly-supervised approach utilizes a tiny number of OOD samples to further enhance the detection performance in various OOD scenarios. Extensive experimental results demonstrate that our proposed methods significantly outperform the baselines in detecting OOD samples from four different scenarios simultaneously and also positively impact a main code understanding task.
Author Yan, Yanfu
Poshyvanyk, Denys
Duong, Viet
Shao, Huajie
Author_xml – sequence: 1
  givenname: Yanfu
  surname: Yan
  fullname: Yan, Yanfu
  email: yyan09@wm.edu
  organization: William & Mary,Department of Computer Science,Williamsburg,Virginia,USA
– sequence: 2
  givenname: Viet
  surname: Duong
  fullname: Duong, Viet
  email: vqduong@wm.edu
  organization: William & Mary,Department of Computer Science,Williamsburg,Virginia,USA
– sequence: 3
  givenname: Huajie
  surname: Shao
  fullname: Shao, Huajie
  email: hshao@wm.edu
  organization: William & Mary,Department of Computer Science,Williamsburg,Virginia,USA
– sequence: 4
  givenname: Denys
  surname: Poshyvanyk
  fullname: Poshyvanyk, Denys
  email: dposhyvanyk@wm.edu
  organization: William & Mary,Department of Computer Science,Williamsburg,Virginia,USA
BookMark eNotkM9OAjEYxKvRREDegENfYPFru_13NAsiCYYDeCbt7re6BrekLSG8vWv0NJPM_OYwY3LXhx4JmTGYMwb2aV3tllKKUs85cDkHYFrfkKnV1gjBJEhl2S0ZMSlNwTiXD2Sc0hcAqNLaEdntw8XFJtG3EJHu4znlS4j580oXiCdahQaHqMFjov5Kl73zx67_oNtzLkJbLLqUY-fPuQv9AGSsf90juW_dMeH0Xyfk_WW5r16LzXa1rp43heMKctGW3iJoro0XwjhwUnmvOfNKgmuYEBxVq2sp0TRcq5IZp71rG6hrzduhMCGzv90OEQ-n2H27eD0Mr3BrBvwHZGNS6Q
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IH
CBEJK
RIE
RIO
DOI 10.1109/ICSE55347.2025.00177
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan (POP) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP) 1998-present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISBN 9798331505691
EISSN 1558-1225
EndPage 781
ExternalDocumentID 11029813
Genre orig-research
GrantInformation_xml – fundername: Cisco Systems
  funderid: 10.13039/100004351
– fundername: NSF
  grantid: CCF-2311469,CNS-2132281,CCF-2007246,CCF-1955853,CNS-2346357
  funderid: 10.13039/100000001
GroupedDBID -~X
.4S
.DC
29O
5VS
6IE
6IF
6IH
6IK
6IL
6IM
6IN
8US
AAJGR
AAWTH
ABLEC
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
ARCSS
AVWKF
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
EDO
FEDTE
I-F
IEGSK
IJVOP
IPLJI
M43
OCL
RIE
RIL
RIO
ID FETCH-LOGICAL-a260t-f4b9e07278b338a0a56bb721b650ad1332e6f7c55e8d276418a7bafd0cc72fd13
IEDL.DBID RIE
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001538318100060&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
IngestDate Wed Aug 27 01:40:13 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a260t-f4b9e07278b338a0a56bb721b650ad1332e6f7c55e8d276418a7bafd0cc72fd13
PageCount 13
ParticipantIDs ieee_primary_11029813
PublicationCentury 2000
PublicationDate 2025-April-26
PublicationDateYYYYMMDD 2025-04-26
PublicationDate_xml – month: 04
  year: 2025
  text: 2025-April-26
  day: 26
PublicationDecade 2020
PublicationTitle Proceedings / International Conference on Software Engineering
PublicationTitleAbbrev ICSE
PublicationYear 2025
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0006499
Score 2.2898738
Snippet Numerous machine learning (ML) models have been developed, including those for software engineering (SE) tasks, under the assumption that training and testing...
SourceID ieee
SourceType Publisher
StartPage 769
SubjectTerms Code Models
Codes
Contrastive learning
Data models
Measurement
OOD detection
Predictive models
Reliability
Software engineering
Testing
Training
Training data
Trustworthy ML
Title Towards More Trustworthy Deep Code Models by Enabling Out-of-Distribution Detection
URI https://ieeexplore.ieee.org/document/11029813
WOSCitedRecordID wos001538318100060&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JSwMxFA5aPHiqS8WdHLzGzmQySebcBQWthVbprWQFQWZKOxX6731Jx4oHD95CyAJ5JO99yfvyIXRXKMO10JQ4nljCPLNEGZERxbxPbGY8i2Iwb09iNJKzWTFuyOqRC-Oci8ln7j4U41u-rcw6XJV1wVXRQgaN2n0h-JastTt2OcTuDTcuTYruY28yyPOMCcCANNybpOK3gkp0IMP2P6c-Qp0fKh4e75zMMdpz5Qlqf2sx4GZrnqLJNOa_rvAzjIyngUkRs_82uO_cAvcq63DQPftYYb3Bg8CYgvHwy7omlSf98H1uo3wFHeqYn1V20OtwMO09kEYwgSiAJTXxTBcugYhEakCeKlE51xognoYwTFlAo9RxL0yeO2mp4CyVSmjlbWKMoB4anKFWWZXuHGGpZJZaTi3gGYCAEFWw3GZ5kViRau_YBeqERZovtn9izL_X5_KP-it0GOwQ3mEov0aterl2N-jAfNbvq-VttOQXPjif3Q
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1JSwMxFA5SBT3VpeJuDl5jZ8kkM-e20mJbCx2lt5IVBJkp7VTov_clTisePHgLIQvkkbz3Je_Lh9BDJhSTXEbEsEATaqkmQvGYCGptoGNlqReDeRvy8TidzbJJTVb3XBhjjE8-M4-u6N_ydanW7qqsDa4qylKnUbvvpLNqutbu4GUQvdfsuDDI2oPOtJckMeWAAiN3cxLy3xoq3oU8Nf85-TFq_ZDx8GTnZk7QnilOUXOrxoDrzXmGprnPgF3hEYyMc8el8Pl_G9w1ZoE7pTbYKZ99rLDc4J7jTMF4-GVdkdKSrvtAt9a-gg6Vz9AqWuj1qZd3-qSWTCACgElFLJWZCSAmSSVgTxGIhEkJIE9CICY04NHIMMtVkphUR5zRMBVcCqsDpXhkocE5ahRlYS4QTkUah5pFGhANgECIK2ii4yQLNA-lNfQStdwizRffv2LMt-tz9Uf9PTrs56PhfDgYP1-jI2cT9yoTsRvUqJZrc4sO1Gf1vlreeat-AcLJoyY
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=proceeding&rft.title=Proceedings+%2F+International+Conference+on+Software+Engineering&rft.atitle=Towards+More+Trustworthy+Deep+Code+Models+by+Enabling+Out-of-Distribution+Detection&rft.au=Yan%2C+Yanfu&rft.au=Duong%2C+Viet&rft.au=Shao%2C+Huajie&rft.au=Poshyvanyk%2C+Denys&rft.date=2025-04-26&rft.pub=IEEE&rft.eissn=1558-1225&rft.spage=769&rft.epage=781&rft_id=info:doi/10.1109%2FICSE55347.2025.00177&rft.externalDocID=11029813