Robust LLM Watermarking with Minimal Semantic Distortion for IP Protection

• URL: http://arxiv.org/abs/2605.23175v1
• Date: Fri, 22 May 2026 02:51:42 GMT
• Title: Robust LLM Watermarking with Minimal Semantic Distortion for IP Protection
• Abstract summary: SAFESEAL is a novel key-conditioned watermarking framework that achieves strong detectability with minimal impact on model utility.<n>We show that SAFESEAL outperforms baselines in utility, detectability, and robustness, achieving a BERTScore of 0.983, entity similarity of 0.963, a 98.2% detection rate, and the highest human ratings for text quality and content, with latency comparable to the fastest baseline.
• Score: 23.7834497390662
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Proprietary large language models (LLMs) face risks of intellectual property (IP) violation, as adversaries can replicate an LLM by collecting input-output pairs to train a surrogate model, causing financial setbacks. Watermarks offer a promising defense to verify ownership, but existing methods often struggle with semantic distortion, factual inconsistency, and adversarial attacks. In addition, key-conditioned watermarks for provider-specific detection, especially in cross-provider and multi-user scenarios, remain largely underexplored. To address these challenges, we propose SAFESEAL, a novel key-conditioned watermarking framework that achieves strong detectability with minimal impact on model utility, effectively balancing detectability, utility, and robustness. SAFESEAL preserves named entities while substituting linguistic terms with context-aware synonyms through a key-conditioned Tournament sampling mechanism, maintaining semantic fidelity and factual consistency. For detection, we introduce a key-conditioned contrastive detector that jointly encodes the text and key, enabling provider-specific and robust watermark verification. We derive theoretical bounds on the utility-detectability trade-off and significantly reduce latency through lightweight models, batching, and parallelism. Extensive experiments show that SAFESEAL outperforms baselines in utility, detectability, and robustness, achieving a BERTScore of 0.983, entity similarity of 0.963, a 98.2% detection rate, and the highest human ratings for text quality and content preservation, with latency comparable to the fastest baseline. To promote transparency and community-driven progress, we release the first public watermark leaderboard and an interactive demo.

Related papers

• PASA: A Principled Embedding-Space Watermarking Approach for LLM-Generated Text under Semantic-Invariant Attacks [3.815798234276803]
  PASA is a principled, robust, and distortion-free watermarking algorithm.<n>It embeds and detects a watermark at the semantic level.
  arXiv  Detail & Related papers  (2026-05-09T01:09:01Z)
• ContractShield: Bridging Semantic-Structural Gaps via Hierarchical Cross-Modal Fusion for Multi-Label Vulnerability Detection in Obfuscated Smart Contracts [0.6936215735722616]
  Smart contracts are increasingly targeted by adversaries using obfuscation techniques to evade vulnerability detection.<n>We propose ContractShield, a robust multimodal framework that correlates multiple complementary features through a three-level fusion.<n>ContractShield simultaneously detects five major vulnerability types with 91 percentage F1-score, outperforming state-of-the-art approaches by 6-15 percentage under adversarial conditions.
  arXiv  Detail & Related papers  (2026-04-03T06:29:34Z)
• A Visual Semantic Adaptive Watermark grounded by Prefix-Tuning for Large Vision-Language Model [48.79816664229285]
  VIsual Semantic Adaptive Watermark (VISA-Mark) is a novel framework that embeds detectable signals while strictly preserving visual fidelity.<n>Our approach employs a lightweight, efficiently trained prefix-tuner to extract dynamic Visual-Evidence Weights.<n> Empirical results confirm that VISA-Mark outperforms conventional methods with a 7.8% improvement in visual consistency.
  arXiv  Detail & Related papers  (2026-01-12T07:55:13Z)
• SWAP: Towards Copyright Auditing of Soft Prompts via Sequential Watermarking [58.475471437150674]
  We propose sequential watermarking for soft prompts (SWAP)<n>SWAP encodes watermarks through a specific order of defender-specified out-of-distribution classes.<n>Experiments on 11 datasets demonstrate SWAP's effectiveness, harmlessness, and robustness against potential adaptive attacks.
  arXiv  Detail & Related papers  (2025-11-05T13:48:48Z)
• DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models [55.30555646945055]
  Text-to-Image (T2I) models are vulnerable to semantic leakage.<n>We introduce DeLeaker, a lightweight approach that mitigates leakage by directly intervening on the model's attention maps.<n>SLIM is the first dataset dedicated to semantic leakage.
  arXiv  Detail & Related papers  (2025-10-16T17:39:21Z)
• An Ensemble Framework for Unbiased Language Model Watermarking [60.99969104552168]
  We propose ENS, a novel ensemble framework that enhances the detectability and robustness of unbiased watermarks.<n>ENS sequentially composes multiple independent watermark instances, each governed by a distinct key, to amplify the watermark signal.<n> Empirical evaluations show that ENS substantially reduces the number of tokens needed for reliable detection and increases resistance to smoothing and paraphrasing attacks.
  arXiv  Detail & Related papers  (2025-09-28T19:37:44Z)
• PMark: Towards Robust and Distortion-free Semantic-level Watermarking with Channel Constraints [49.2373408329323]
  We introduce a new theoretical framework on watermark-leveling (SWM) for large language models (LLMs)<n>We propose PMark, a simple yet powerful SWM method that estimates the median next sentence dynamically through sampling channels.<n> Experimental results show that PMark consistently outperforms existing SWM baselines in both text quality and paraphrasing.
  arXiv  Detail & Related papers  (2025-09-25T12:08:31Z)
• StableGuard: Towards Unified Copyright Protection and Tamper Localization in Latent Diffusion Models [55.05404953041403]
  We propose a novel framework that seamlessly integrates a binary watermark into the diffusion generation process.<n>We show that StableGuard consistently outperforms state-of-the-art methods in image fidelity, watermark verification, and tampering localization.
  arXiv  Detail & Related papers  (2025-09-22T16:35:19Z)
• Semantic Watermarking Reinvented: Enhancing Robustness and Generation Quality with Fourier Integrity [31.666430190864947]
  We propose a novel embedding method called Hermitian Symmetric Fourier Watermarking (SFW)<n>SFW maintains frequency integrity by enforcing Hermitian symmetry.<n>We introduce a center-aware embedding strategy that reduces the vulnerability of semantic watermarking due to cropping attacks.
  arXiv  Detail & Related papers  (2025-09-09T12:15:16Z)
• Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models via Reasoning [58.57194301645823]
  Large language models (LLMs) are increasingly integrated into real-world personalized applications.<n>The valuable and often proprietary nature of the knowledge bases used in RAG introduces the risk of unauthorized usage by adversaries.<n>Existing methods that can be generalized as watermarking techniques to protect these knowledge bases typically involve poisoning or backdoor attacks.<n>We propose name for harmless' copyright protection of knowledge bases.
  arXiv  Detail & Related papers  (2025-02-10T09:15:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.