LLaMA-Reviewer: Advancing Code Review Automation with Large Language Models through Parameter-Efficient Fine-Tuning

The automation of code review activities, a long-standing pursuit in software engineering, has been primarily addressed by numerous domain-specific pre-trained models. Despite their success, these models frequently demand extensive resources for pre-training from scratch. In contrast, Large Language...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings - International Symposium on Software Reliability Engineering s. 647 - 658
Hlavní autoři: Lu, Junyi, Yu, Lei, Li, Xiaojia, Yang, Li, Zuo, Chun
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 09.10.2023
Témata:
ISSN:2332-6549
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract The automation of code review activities, a long-standing pursuit in software engineering, has been primarily addressed by numerous domain-specific pre-trained models. Despite their success, these models frequently demand extensive resources for pre-training from scratch. In contrast, Large Language Models (LLMs) provide an intriguing alternative, given their remarkable capabilities when supplemented with domain-specific knowledge. However, their potential for automating code review tasks remains largely unexplored.In response to this research gap, we present LLaMA-Reviewer, an innovative framework that leverages the capabilities of LLaMA, a popular LLM, in the realm of code review. Mindful of resource constraints, this framework employs parameter-efficient fine-tuning (PEFT) methods, delivering high performance while using less than 1% of trainable parameters.An extensive evaluation of LLaMA-Reviewer is conducted on two diverse, publicly available datasets. Notably, even with the smallest LLaMA base model consisting of 6.7B parameters and a limited number of tuning epochs, LLaMA-Reviewer equals the performance of existing code-review-focused models.The ablation experiments provide insights into the influence of various fine-tuning process components, including input representation, instruction tuning, and different PEFT methods. To foster continuous progress in this field, the code and all PEFT-weight plugins have been made open-source.
AbstractList The automation of code review activities, a long-standing pursuit in software engineering, has been primarily addressed by numerous domain-specific pre-trained models. Despite their success, these models frequently demand extensive resources for pre-training from scratch. In contrast, Large Language Models (LLMs) provide an intriguing alternative, given their remarkable capabilities when supplemented with domain-specific knowledge. However, their potential for automating code review tasks remains largely unexplored.In response to this research gap, we present LLaMA-Reviewer, an innovative framework that leverages the capabilities of LLaMA, a popular LLM, in the realm of code review. Mindful of resource constraints, this framework employs parameter-efficient fine-tuning (PEFT) methods, delivering high performance while using less than 1% of trainable parameters.An extensive evaluation of LLaMA-Reviewer is conducted on two diverse, publicly available datasets. Notably, even with the smallest LLaMA base model consisting of 6.7B parameters and a limited number of tuning epochs, LLaMA-Reviewer equals the performance of existing code-review-focused models.The ablation experiments provide insights into the influence of various fine-tuning process components, including input representation, instruction tuning, and different PEFT methods. To foster continuous progress in this field, the code and all PEFT-weight plugins have been made open-source.
Author Yang, Li
Zuo, Chun
Li, Xiaojia
Yu, Lei
Lu, Junyi
Author_xml – sequence: 1
  givenname: Junyi
  surname: Lu
  fullname: Lu, Junyi
  email: lujunyi21@mails.ucas.ac.cn
  organization: Institute of Software,Chinese Academy of Sciences,Beijing,China
– sequence: 2
  givenname: Lei
  surname: Yu
  fullname: Yu, Lei
  email: yulei21@mails.ucas.ac.cn
  organization: Institute of Software,Chinese Academy of Sciences,Beijing,China
– sequence: 3
  givenname: Xiaojia
  surname: Li
  fullname: Li, Xiaojia
  email: lixj21@mails.tsinghua.edu.cn
  organization: Tsinghua University,School of Software,Beijing,China
– sequence: 4
  givenname: Li
  surname: Yang
  fullname: Yang, Li
  email: yangli2017@iscas.ac.cn
  organization: Institute of Software,Chinese Academy of Sciences,Beijing,China
– sequence: 5
  givenname: Chun
  surname: Zuo
  fullname: Zuo, Chun
  email: zuochun@sinosoft.com.cn
  organization: Sinosoft Company Limited,Beijing,China
BookMark eNotjttKAzEYhKMoWGvfQCEvkJrDnuLdUlotbFHa3pe_mz_bSJuVbLbFt3eh3swMzMcwj-TOtx4JeRF8KgTXr8vNZj1PdZEUU8mlmnLOZXZDJjrXhUq5EqlO1C0ZSaUky9JEP5DHrvseKJ4IOSJdVcGqZGs8O7xgeKOlOYOvnW_orDVIrwUt-9ieILrW04uLB1pBaHBQ3_QwhNWAHjsaD6HtmwP9ggAnjBjY3FpXO_SRLpxHtu39sPxE7i0cO5z8-5hsF_Pt7INVn-_LWVkxN5yLTGizVypTKGuZYZJnAmwO3IKQyqQq0xpB1JlNc4PG2LrmSVELC-k-2RtZqDF5vs46RNz9BHeC8LsTXGqtVaH-ANkLXu8
CODEN IEEPAD
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ISSRE59848.2023.00026
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Xplore POP ALL
IEEE Xplore All Conference Proceedings
IEEE/IET Electronic Library (IEL) (UW System Shared)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
EISBN 9798350315943
EISSN 2332-6549
EndPage 658
ExternalDocumentID 10299938
Genre orig-research
GrantInformation_xml – fundername: Science and Technology Service Network Plan
  funderid: 10.13039/501100013315
GroupedDBID 6IE
6IF
6IH
6IK
6IL
6IN
AAJGR
AAWTH
ABLEC
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
IEGSK
IPLJI
M43
OCL
RIE
RIL
RNS
ID FETCH-LOGICAL-i204t-19db3363e2c26e4761af7a0fa123d53699ea1c6f57deddfcc048c1fa5b4bd283
IEDL.DBID RIE
ISICitedReferencesCount 54
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001096886300057&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 02:30:37 EDT 2025
IsPeerReviewed false
IsScholarly true
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-i204t-19db3363e2c26e4761af7a0fa123d53699ea1c6f57deddfcc048c1fa5b4bd283
PageCount 12
ParticipantIDs ieee_primary_10299938
PublicationCentury 2000
PublicationDate 2023-Oct.-9
PublicationDateYYYYMMDD 2023-10-09
PublicationDate_xml – month: 10
  year: 2023
  text: 2023-Oct.-9
  day: 09
PublicationDecade 2020
PublicationTitle Proceedings - International Symposium on Software Reliability Engineering
PublicationTitleAbbrev ISSRE
PublicationYear 2023
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0020412
Score 2.5704036
Snippet The automation of code review activities, a long-standing pursuit in software engineering, has been primarily addressed by numerous domain-specific pre-trained...
SourceID ieee
SourceType Publisher
StartPage 647
SubjectTerms Automation
Code Review Automation
Codes
Deep Learning
Large Language Models (LLMs)
LLaMA
Parameter-Efficient Fine-Tuning (PEFT)
Quality assurance
Software engineering
Software Quality Assurance
Software reliability
Task analysis
Tuning
Title LLaMA-Reviewer: Advancing Code Review Automation with Large Language Models through Parameter-Efficient Fine-Tuning
URI https://ieeexplore.ieee.org/document/10299938
WOSCitedRecordID wos001096886300057&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/eLvHCXMwlV07a8MwEBZN6NApfaT0jYauam3ZlqxuISS04IbQhJIt6HGCQIlLYvf3V5KdtEuHLsbIGJkT59N9uu8-hO65SxEU9WycmBmSugyFyNxYwigFyF3AhoDpvhd8MskXCzFtyeqBCwMAofgMHvxtOMs3pa49VOY83P08RZJ3UIdz1pC19tmVbxzVUnTiSDy-zGZvo0zkoX6L-jamoYHCLwmVEEHGvX_OfYz6P1w8PN1HmRN0AOtT1NuJMeDWN8_Qtijk64A0WD9snnDQS9buHTwsDeDmAR7UVdmwFbGHYHHhK8HdtUEtsZdG-9jiVr0HT6Wv3XLzkFFoNeE-EY_dvpTMa4-n9NF8PJoPn0mrqEBWzj4ViYVRScISoJoySDmLpeUystLFL5MlTAiQsWY24waMsVo7_9axlZlKlXEbkXPUXZdruEA4l4qqyCoe6TyFJFXUMpmolLoBrTm7RH1vw-Vn0zNjuTPf1R_j1-jIL1MokxM3qFttarhFh_qrWm03d2GlvwHqGKud
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV3NS8MwFA86BT3Nj4nf5uA12qZt2ngbY2PDbgxXZLeRj1cYyCpb699vknbTiwcvpaSUlBdeX94v7_d-CD3GJkWQ1LJxfKZJaDIUIhKdE0YpQGICNjhM9z2NJ5NkPufThqzuuDAA4IrP4MneurN8XajKQmXGw83PkwfJPjqIwpB6NV1rl1_Z1lENScf3-PNoNnvrRzxxFVzUNjJ1LRR-iai4GDJo_3P2E9T5YePh6S7OnKI9WJ2h9laOATfeeY42aSrGXVKj_bB-wU4xWZl3cK_QgOsHuFuVRc1XxBaExamtBTfXGrfEVhztY4Mb_R48FbZ6y8xD-q7ZhPlEPDA7U5JVFlHpoGzQz3pD0mgqkKWxT0l8rmUQsACoogzCmPkij4WXCxPBdBQwzkH4iuVRrEHrXCnj4crPRSRDqc1W5AK1VsUKLhFOhKTSy2XsqSSEIJQ0ZyKQITUDSsXsCnWsDRefddeMxdZ813-MP6CjYTZOF-lo8nqDju2SuaI5fota5bqCO3SovsrlZn3vVv0bARau5A
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+-+International+Symposium+on+Software+Reliability+Engineering&rft.atitle=LLaMA-Reviewer%3A+Advancing+Code+Review+Automation+with+Large+Language+Models+through+Parameter-Efficient+Fine-Tuning&rft.au=Lu%2C+Junyi&rft.au=Yu%2C+Lei&rft.au=Li%2C+Xiaojia&rft.au=Yang%2C+Li&rft.date=2023-10-09&rft.pub=IEEE&rft.eissn=2332-6549&rft.spage=647&rft.epage=658&rft_id=info:doi/10.1109%2FISSRE59848.2023.00026&rft.externalDocID=10299938