From Brittle LLM Code Reasoning to MaxSAT-Based Verified Repairs @ UCL
In this talk, we examine the limitations of Large Language Models (LLMs) in semantic code reasoning, showing that their predictions may change under semantics-preserving code …
Towards Assessing and Repairing LLM-Generated Code via Model Checking and MaxSAT-Based Fault Localisation @ Dagstuhl 2026
LLMs for code often lack true semantic understanding, evidenced by their instability under semantics-preserving transformations, and we address this by integrating formal methods …
PyVeritas: On Verifying Python via LLM-Based Transpilation and Bounded Model Checking for C @ P-AI-FM@AAAI 2026
In this talk I will present PyVeritas, a novel framework that leverages Large Language Models (LLMs) for high-level transpilation from Python to C, followed by bounded model …
PyVeritas: On Verifying Python via LLM-Based Transpilation and Bounded Model Checking for C
In this paper, we propose PyVeritas, a novel framework that leverages Large Language Models (LLMs) for high-level transpilation from Python to C, followed by bounded model checking …
📄 Paper accepted @ the Post-AI Formal Methods Workshop @ AAAI 2026! 🎉
I’m excited to share that our paper on PyVeritas has been accepted to the P-AI-FM-26 Workshop @ AAAI 2026! 📄
MENTOR: Automated Feedback for Introductory Programming Exercises
This PhD thesis presents MENTOR, a semantic automated program repair (APR) framework designed to provide Automated Feedback for Introductory Programming Exercises.
Counterexample Guided Program Repair Using Zero-Shot Learning and MaxSAT-based Fault Localization
In this paper, we propose a novel approach that combines the strengths of both FM-based fault localization and LLMs, via zero-shot learning, to enhance APR for IPAs. Our method …
Model-Based Diagnosis for Software
Localising system faults has long been recognised as one of the most time-consuming and costly tasks in software engineering. Given a buggy system, fault localisation (FL) refers …
📄 Paper accepted @ AAAI 2025!! 🎉
I am very happy to share that our paper that combines the strengths of both MaxSAT-based fault localisation and Large Language Models, via zero-shot learning, to enhance automated …
Automated Feedback for Introductory Programming Exercises
Delivering valuable and personalised feedback to students remains one of the greatest challenges in programming education, particularly in courses with large enrollments. Providing …
CFaults: Model-Based Diagnosis for Fault Localization in C with Multiple Test Cases @ FM 2024
In this talk I will introduce a novel fault localization approach for C programs with multiple faults. CFaults leverages Model-Based Diagnosis (MBD) with multiple observations and …
CFaults: Model-Based Diagnosis for Fault Localization in C with Multiple Test Cases
This paper introduces a novel fault localization approach for C programs with multiple faults. CFaults leverages Model-Based Diagnosis (MBD) with multiple observations and …
User partitioning for Maximum Satisfiability
The Maximum Satisfiability (MaxSAT) problem is the optimisation variant of the Satisfiability (SAT) problem. Solving MaxSAT efficiently often involves partitioning the set of soft …
UpMax: User partitioning for MaxSAT @ SAT 2023
In this talk I will present a new framework called UpMax that decouples the partitioning procedure from the MaxSAT solving algorithms.
UpMax: User partitioning for MaxSAT
This paper proposes a new framework called UpMax that decouples the partitioning procedure from the MaxSAT solving algorithms. As a result, new partitioning procedures can be …
AlloyMax: Bringing Maximum Satisfaction to Relational Specifications
We propose AlloyMax, an extension of Alloy with a capability to express and analyze problems with optimal solutions. AlloyMax introduces (1) a small addition of language constructs …
Changjian Zhang
•
•
1 min read