Proposes and open-sources an automated framework for species-fair security comparisons of human-written, LLM-generated, and hybrid code, validated via feasibility study.
Firestone, Performance vs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
How to Compare the Security of Code Written by Humans to LLM-generated Code
Proposes and open-sources an automated framework for species-fair security comparisons of human-written, LLM-generated, and hybrid code, validated via feasibility study.