Replication Package: Do LLMs Understand Compiler Errors? A Compiler-Grounded Study of First-Diagnostic Control

Read full text Summarise with VIRAT
Authors: Anonymous
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Replication package for the ASE 2026 submission "Do LLMs Understand Compiler Errors? A Compiler-Grounded Study of First-Diagnostic Control." Contents:- 12,800 raw LLM outputs (8 models × 8 error types × 2 languages × 100 samples)- MSVP verification pipeline (deterministic compiler-based classification)- Prompt templates (generation, recognition, IBP, CDP, refinement)- Baseline results (rule-based mutation, 1-shot, chain-of-thought)- Cross-compiler analysis (gcc-14 vs clang-18)- Expert judgment survey data- All analysis scripts for reproducing tables and figures See README.md for setup and reproduction instructions.

Fields of Study & Subjects

Software Engineering Research Software Testing and Debugging Techniques Logic programming and type systems

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
27 March 2026
Language
EN
Document type
Dataset