Negar Fathi

Ph.D. Candidate in Computer Science at University of Nebraska-Lincoln

About Me

I am a Ph.D. candidate in Computer Science at the University of Nebraska–Lincoln. My research focuses on formal methods, program analysis, and software testing to improve software reliability and correctness. My work also explores how artificial intelligence, including large language models, can be combined with rigorous program reasoning to support software engineering.

Featured Publications

Sound Termination and Non-termination Analysis of C Programs with Bit-Precise Bounded Semantics and Advanced Constructs

Conference

Negar Fathi, Hiroshi Unno, Tachio Terauchi, and Rahul Purandare

Proceedings of the ACM on Software Engineering, FSE

2026

DOI Code Video

COSMOS: A Comprehensive Framework for Automatically Generating Domain-Oriented Test Suite

Journal

Akram Kalaee, Saeed Parsa, and Negar Fathi

Information and Software Technology

2023

DOI Code

Loop-Based Slicing and Input-Driven Concretization: An Empirical Study of Termination and Non-Termination Analysis

Preprint

Negar Fathi, Rahul Purandare, Tachio Terauchi, and Hiroshi Unno

arXiv preprint; under review at IEEE Transactions on Software Engineering

2026

DOI Code

View All Publications

Featured Projects

Nexus

in progress

Apr. 2026 - Present

Developed a neuro-symbolic framework for C program termination and non-termination analysis, integrating LLVM-based loop extraction, LLM-guided witness synthesis, and SMT-based validation with iterative refinement.

GitHub

Athena

completed

Aug. 2024 - Sep. 2025

Developed a sound framework for termination and non-termination analysis of C programs, incorporating bit-precise reasoning, bounded semantics, and support for complex language constructs.

GitHub

EDFNet

completed

Aug. 2025 - Dec. 2025

Developed and evaluated an early-fusion semantic segmentation framework integrating RGB, depth, and edge information for thin-obstacle detection in UAV navigation using the DDOS dataset.

GitHub

View All Projects