RoCoMark: Robust Code Watermarking Against LLM-Based Functionality-Preserving Rewriting Attacks

Summary

RoCoMark is a code watermarking framework designed to survive LLM-based rewriting that preserves functionality. The authors attribute existing methods' failures to a distribution mismatch between training perturbations and LLM rewriting behavior. Across four benchmarks and attacks from five LLMs, RoCoMark reaches an average detection AUROC of 74.2% and a Rewrite BitACC of 75.3%, with a code pass rate above 99.7%.

Related items

• HighCVE-2026-123818: LMDeploy is a toolkit for compressing, deploying, and serving large language models. From 0.12.3 until 0.15.0…Similar attack · NVD/CVE Database
• HighGHSA-vvr2-h2jp-838m: Code Graph RAG: MCP paginated read_file path traversal / arbitrary file readSimilar attack · GitHub Advisory Database
• InfoSuppression Is Not Erasure: Mechanistic Analysis and Reactivation Boundaries of Subspace Projection for LoRA Backdoor DefenseSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• MediumAttackers Crash Claude's 'Office Party' Flaw to Run Malicious Code on LinuxSimilar attack · Dark Reading
• InfoSay Shibboleth: Fingerprinting the LLMs and Harnesses Behind Agentic AttacksSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)