Reproduce “NOIR: Privacy-Preserving Generation of Code with Open-Source LLMs” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Khoa Nguyen, Khiem Ton, NhatHai Phan, Issa Khalil, Khang Tran, Cristian Borcea, Ruoming Jin, Abdallah Khreishah and 1 more

NOIR splits an open-source model between client and cloud for private code generation. The client randomizes prompt representations and remaps token indices; the cloud processes the middle portion, and the client generates code locally. On CodeQwen1.5-7B-Chat, BigCodeBench “Instruct” Pass@1 was 31.1 versus 31.7 for the clean model at ε=27, with zero reported PromptRA and CodeRA attack-success rates under tested conditions.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.