Preserving Structure, Strengthening Signals: A Grammar-Driven Approach to Optimize the Trade-off Between Quality and Detectability in Code Watermarking
cs.CR, cs.CL
Submitted: 2026-07-11
Updated: 2026-10-12
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Evaluating Large Language Models Trained on Code
- BiMark: Unbiased Multilayer Watermarking for Large Language Models
- Qwen2.5-Coder Technical Report
- Token-Specific Watermarking with Enhanced Detectability and Semantic Coherence for Large Language Models
- Marking Code Without Breaking It: Code Watermarking for Detecting LLM-Generated Code
- On the Reliability of Watermarks for Large Language Models
- Turning Your Strength into Watermark: Watermarking Large Language Model via Knowledge Injection
- StarCoder 2 and The Stack v2: The Next Generation
- An Entropy-based Text Watermarking Detection Method
- Watermarking Low-entropy Generation for Large Language Models: An Unbiased and Low-risk Method
- OctoPack: Instruction Tuning Code Large Language Models
- Embarrassingly Simple Text Watermarks
- Is The Watermarking Of LLM-Generated Code Robust?
- Necessary and Sufficient Watermark for Large Language Models
- Towards Codable Watermarking for Injecting Multi-bits Information to LLMs
- Qwen3 Technical Report
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Estimating the Social Cost of Corporate Data Breaches
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills