Is Watermarking LLM-Generated Code Robust?

arXiv:2403.17983v1 Announce Type: cross
Abstract: We present the first study of the robustness of existing watermarking techniques on Python code generated by large language models. Although existing works showed that watermarking can be robust for natural language, we show that it is easy to remove these watermarks on code by semantic-preserving transformations.

abstract arxiv code cs.cr cs.lg easy generated language language models large language large language models llm natural natural language python robust robustness semantic show study type watermarking watermarks