Ming Wen

From Function to Repository: Towards Understanding and an Agentic Approach for Repo-Level Directed Input Generation

AgenticDIG is designed, an agentic workflow that integrates target-oriented analysis, intention extraction, fault injection, fault injection, and iterative feedback to improve repository-level DIG.

Testing Static Analyzers via Semantic-Preserving Mutators Learned from Real-World Refactoring Practice

STARS: Static Analysis-Guided Assertion Synthesis using Large Language Models

Automated unit test generation promises to reduce the cost of software quality assurance, and hence, is attracting attention from both academia and industry. Yet, generating assertions that are executable, meaningful to developers, and able to catch faults remains an unsolved challenge. Existing approaches either rando...

Validating Rust Compilers with Trait-Type Constraint Graph

This work proposes a validation methodology based on a novel concept of Trait-Type Constraint Graph (TTCG), a graph that captures semantic relationships between traits and types and enables the derivation of the Trait Constraint Space (TCS) for systematic exploration.

Understanding and Detecting Android R8 Compiler Bugs

An automated testing tool named R8Scan is developed that utilizes a novel idea to synthesize seeds from prioritized real-world functions and construct the corresponding arguments empowered by Large Language Models to test R8, thus enabling the exploration of a broader range of semantics.