ACL Findings2026
SLIM: Stealthy Low-Coverage Black-Box Watermarking via Latent-Space Confusion Zones
Hengyu Wu, Yang Cao
Abstract
Training data is a critical and often proprietary asset in Large Language Model (LLM) development, motivating the use of data watermarking to embed model-transferable signals for usage verification. We identify low coverage as a vital yet largely overlooked requirement for practicality, as individual data owners typically contribute only a minute fraction of massive training corpora. Prior methods fail to maintain stealthiness, verification feasibility, or robustness when only one or a few sequences can be modified. To address these limitations, we introduce SLIM, a framework enabling per-user data provenance verification under strict blackbox access. SLIM leverages intrinsic LLM properties to induce a Latent-Space Confusion Zone by training the model to map semantically similar prefixes to divergent continuations. This manifests as localized generation instability, which can be reliably detected via hypothesis testing. Experiments demonstrate that SLIM achieves ultra-low coverage capability, strong black-box verification performance, and great scalability while preserving both stealthiness and model utility, offering a robust solution for protecting training data in modern LLM pipelines 1 .
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on27
- Language Models are Few-Shot LearnersTom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah et al.NeurIPS 2020 · 64,255 citations
- Training language models to follow instructions with human feedbackLong Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida et al.NeurIPS 2022 · 24,707 citations
- BERTScore: Evaluating Text Generation with BERTTianyi Zhang, Varsha Kishore, Felix Wu, Kilian Q. Weinberger et al.ICLR 2020 · 8,443 citations
- Measuring Massive Multitask Language UnderstandingDan Hendrycks, Collin Burns, Steven Basart, Andy Zou et al.ICLR 2021 · 7,905 citations
- Deep Learning with Differential PrivacyMartín Abadi, Andy Chu, Ian J. Goodfellow, H. Brendan McMahan et al.CCS 2016 · 7,620 citations
Related papers
- Ghost in the Transformer: Detecting Model Reuse with Invariant Spectral SignaturesSuqing Wang, Ziyang Ma, Xinyi Li, Zuchao LiAAAI 2026 · 1 citation
- StealthInk: A Multi-bit and Stealthy Watermark for Large Language ModelsYa Jiang, Chuxiong Wu, Massieh Kordi Boroujeny, Brian L. Mark et al.ICML 2025
- WASA: WAtermark-based Source Attribution for Large Language Model-Generated DataXinyang Lu, Jingtan Wang, Zitong Zhao, Zhongxiang Dai et al.ACL Findings 2025
- Logit Laundering: Evading Data-Use Auditing in Large Language ModelsRuihan Hu, Wei Luo, Yu-Ming Shang, Jiakai Wang et al.CCS 2026
- LLM Fingerprinting via Semantically Conditioned WatermarksThibaud Gloaguen, Robin Staab, Nikola Jovanovic, Martin T. VechevICLR 2026 · 8 citations
