LineageMark：面向模型派生链的多用户白盒水印贡献追踪框架（已撤稿）
LineageMark: Multi-user White-box Watermarking for Contribution Tracing in Model Derivation Chains
AI 导读
arXiv 论文提出 LineageMark，一种用于大语言模型多阶段派生链的多用户白盒水印框架，通过投影方式在模型参数中编码水印，选用稳定载体并以边际约束维持信号完整性。实验显示其可在多阶段派生中保留贡献者水印，支持增量多用户插入，并对再水印、微调、量化与剪枝具有鲁棒性。作者已因研究大幅修订而撤回该稿件。
正文 · 原文
This paper has been withdrawn by Xiaofeng Xu
No PDF available, click to view other formats
Abstract:In open large language model (LLM) ecosystems, models are frequently adapted across multiple domains and applications, forming multi-stage derivation chains. Consequently, tracking and verifying historical contributions is essential for model provenance and intellectual property protection. However, existing watermarking methods are mainly designed for single-user, one-time embeddings, often fail under repeated model derivation and incremental updates. To address this problem, we propose LineageMark, a multi-user white-box watermarking framework for model derivation chains. The framework encodes watermarks in model parameters using a projection-based approach. Stable carriers are first selected to reduce sensitivity to model changes, each watermark bit is then represented as a projection statistic over these carriers. Additional watermark insertions introduce only bounded perturbations in the projection space, and margin constraints are used to maintain signal integrity. We evaluate the effectiveness of LineageMark in multi-stage model derivation chains. Experimental results show that LineageMark preserves contributor watermarks across multi-stage derivation and supports incremental multi-user watermark insertion. Furthermore, it exhibits robustness against perturbations such as re-watermarking, fine-tuning, quantization, and pruning.
| Comments: | The authors have decided to withdraw this manuscript because the research has undergone substantial revisions in its problem formulation, scope, and experimental evaluation. As a result, the current version no longer adequately represents the revised study. We therefore withdraw this earlier version to avoid potential misunderstandings regarding the research findings |
| Subjects: | Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2606.17123 [cs.CR] |
| (or arXiv:2606.17123v2 [cs.CR] for this version) | |
| https://doi.org/10.48550/arXiv.2606.17123 arXiv-issued DOI via DataCite |
Submission history
From: Xiaofeng Xu [view email]
[v1]
Mon, 15 Jun 2026 13:43:43 UTC (371 KB)
[v2]
Tue, 27 Oct 2026 14:37:50 UTC (1 KB) (withdrawn)
来源：arXiv cs.AI · arxiv.org