Publications

Can LLMs Reason About Program Semantics? A Comprehensive Evaluation of LLMs on Formal Specification Inference

Thanh Le-Cong, Bach Le, Toby Murray

TL;DR: FormalBench: A dataset for benchmarking LLMs' reasoning on program semantics via formal specification inference

ACL 2025, Main Conference

Toward Realistic Evaluations of Just-In-Time Vulnerability Prediction

Duong Nguyen, Thanh Le-Cong, Triet Huynh Minh Le, M. Ali Babar, Quyet-Thang Huynh

TL;DR: Revisit the effectiveness of Just-In-Time Vulnerability Prediction techniques with realistic evaluation settings.

ICSME 2025, Research Track [Direct Accepted: 6.5%]

Towards Reliable Evaluation of Neural Program Repair with Natural Robustness Testing

Thanh Le-Cong, Dat Nguyen, Bach Le, Toby Murray

TL;DR: An empirical study on the robustness of Neural Program Repair techniques against natural semantic-preserving transformations.

TOSEM, Just Accepted, 2025

LEGION: Harnessing Pre-trained Language Models for GitHub Topic Recommendations with Distribution-Balance Loss

Yen-Trang Dang, Thanh Le-Cong, Phuc-Thanh Nguyen, Anh M. T. Bui, Phuong T. Nguyen, Bach Le, and Quyet-Thang Huynh

TL;DR: Investigating and addressing the impact of Long-tailed distribution of GitHub topics on the performance of pre-trained language models for topic recommendation.

EASE 2024, Research Track [Acceptance Rate: 20%]

Leveraging Large Language Model for Automatic Patch Correctness Assessment

Xin Zhou, Bowen Xu, Kisub Kim, DongGyun Han, Hung Nguyen, Thanh Le-Cong, Junda He, Bach Le, David Lo

TL;DR: Automatic Patch Correctness Assessment in Program Repair with Large Language Models.

TSE Volume 50, 2024

Refining ChatGPT-Generated Code: Characterizing and Mitigating Code Quality Issues?

Yue Liu, Thanh Le-Cong, Ratnadira Widyasari, Chakkrit Tantithamthavorn, Li Li, Bach Le, David Lo

TL;DR: An empirical study on code quality issues in ChatGPT-generated code.

TOSEM Volume 33, 2024

MiDas: Multi-Granularity Detector for Vulnerability Fixes

Truong Giang Nguyen, Thanh Le-Cong, Hong Jin Kang, Ratnadira Widyasari, Chengran Yang, Zhipeng Zhao, Bowen Xu, Jiayuan Zhou, Xin Xia, Ahmed E. Hassan, Bach Le, David Lo

TL;DR: Identifying vulnerability fixes by analyzing multi-granularity of code changes.

TSE Volume 49, 2023

Invalidator: Automated Patch Correctness Assessment via Semantic and Syntactic Reasoning

Thanh Le-Cong, Duc-Minh Luong, Bach Le, David Lo, Nhat Hoa Tran, Quang Huy Bui, Quyet Thang Huynh

TL;DR: Reasoning about the correctness of APR-generated patches via program invariants and code representation learning.

TSE Volume 49, 2023

Chronos: Time-Aware Zero-Shot Identification of Libraries from Vulnerability Reports

Yunbo Lyu⁺, Thanh Le-Cong⁺, Hong Jin Kang, Ratnadira Widyasari, Zhao Zhipeng, Bach Le, Ming Li, David Lo

TL;DR: Identifying vulnerable libraries from vulnerability reports via zero-shot learning and domain-specific mechanisms.

ICSE 2023, Research Track [Acceptance Rate: 26%]

Topic Recommendation for GitHub Repositories: How Far Can Extreme Multi-Label Learning Go?

Ratnadira Widyasari, Zhipeng Zhao, Thanh Le Cong, Hong Jin Kang, David Lo

TL;DR: Empirical study on the effectiveness of Extreme Multi-Label Learning and existing techniques for GitHub topic recommendation in a realistic setting.

SANER 2023, Research Track [Acceptance Rate: 27%]

FFL: Fine-grained Fault Localization for Student Programs via Syntactic and Semantic Reasoning

Thanh-Dat Nguyen, Thanh Le-Cong, Duc-Minh Luong, Van-Hai Duong, Bach Le, David Lo, Quyet-Thang Huynh

TL;DR: Automatically identifying fault locations in student programs by applying Graph Neural Network on a fine-grained graph-based representation of the program, which combines AST with test coverage information.

ICSME 2022, Research Track [Acceptance Rate: 23%]

AutoPruner: Transformer-Based Call Graph Pruning

Thanh Le-Cong, Hong Jin Kang, Truong Giang Nguyen, Stefanus Agus Haryono, David Lo, Bach Le, Thang Huynh Quyet

TL;DR: Pruning false positives in static call graph via code features learned by Large Language Model and syntactic features extracted from the original call graph.

FSE 2022, Research Track [Acceptance Rate: 22%]

Usability and Aesthetics: Better Together for Automated Repair of Web Pages

Thanh Le-Cong, Bach Le, Quyet-Thang Huynh, Phi Le Nguyen

TL;DR: Automatically repairing mobile-unfriendly web pages using Evolutionary Optimization.

IEEE 32nd International Symposium on Software Reliability Engineering (ISSRE 2021), Research Track [Acceptance Rate: 27%]