CodeIP: A Grammar-Guided Multi-Bit Watermark for Large Language Models of Code
This addresses the need to protect intellectual property and prevent cheating in programming by enabling provenance tracking for AI-generated code, though it is an incremental improvement over single-bit methods.
The paper tackles the problem of identifying AI-generated code and its source model by introducing CodeIP, a multi-bit watermarking technique that embeds vendor IDs into generated code, achieving effective watermarking across five programming languages while maintaining syntactical correctness.
Large Language Models (LLMs) have achieved remarkable progress in code generation. It now becomes crucial to identify whether the code is AI-generated and to determine the specific model used, particularly for purposes such as protecting Intellectual Property (IP) in industry and preventing cheating in programming exercises. To this end, several attempts have been made to insert watermarks into machine-generated code. However, existing approaches are limited to inserting only a single bit of information. In this paper, we introduce CodeIP, a novel multi-bit watermarking technique that inserts additional information to preserve crucial provenance details, such as the vendor ID of an LLM, thereby safeguarding the IPs of LLMs in code generation. Furthermore, to ensure the syntactical correctness of the generated code, we propose constraining the sampling process for predicting the next token by training a type predictor. Experiments conducted on a real-world dataset across five programming languages demonstrate the effectiveness of CodeIP in watermarking LLMs for code generation while maintaining the syntactical correctness of code.