Cover Image

Watermarking Degrades Alignment in Language Models: Analysis and Mitigation

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Verma, Apurv, Phan, NhatHai, Trivedi, Shubhendu
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Cryptography and Security Machine Learning I.2.7
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Operationalizing a Threat Model for Red-Teaming Large Language Models (LLMs)
by: Verma, Apurv, et al.
Published: (2024)

sudoLLM: On Multi-role Alignment of Language Models
by: Saha, Soumadeep, et al.
Published: (2025)

GuardVal: Dynamic Large Language Model Jailbreak Evaluation for Comprehensive Safety Testing
by: Zhang, Peiyan, et al.
Published: (2025)

Ignore Me But Don't Replace Me: Utilizing Non-Linguistic Elements for Pretraining on the Cybersecurity Domain
by: Jang, Eugene, et al.
Published: (2024)

Prompted Contextual Vectors for Spear-Phishing Detection
by: Nahmias, Daniel, et al.
Published: (2024)

A Semantic Invariant Robust Watermark for Large Language Models
by: Liu, Aiwei, et al.
Published: (2023)

Scaling Trends in Language Model Robustness
by: Howe, Nikolaus, et al.
Published: (2024)

Adversarial Attacks on Large Language Models Using Regularized Relaxation
by: Chacko, Samuel Jacob, et al.
Published: (2024)

Unlocking the Potential of Large Language Models for Clinical Text Anonymization: A Comparative Study
by: Pissarra, David, et al.
Published: (2024)

Seeing the Forest through the Trees: Data Leakage from Partial Transformer Gradients
by: Li, Weijun, et al.
Published: (2024)

Monotonicity as an Architectural Bias for Robust Language Models
by: Cooper, Patrick, et al.
Published: (2026)

MarkLLM: An Open-Source Toolkit for LLM Watermarking
by: Pan, Leyi, et al.
Published: (2024)

Can Watermarked LLMs be Identified by Users via Crafted Prompts?
by: Liu, Aiwei, et al.
Published: (2024)

Same Payload, Different Channel: Measuring Trust Asymmetry in Tool-Using Language Models
by: Syed, Mohammed Sameer, et al.
Published: (2026)

Protection Is (Nearly) All You Need: Structural Protection Dominates Scoring in Globally Capped KV Eviction
by: Garcia, Gabriel
Published: (2026)

PoTS: Proof-of-Training-Steps for Backdoor Detection in Large Language Models
by: Seddik, Issam, et al.
Published: (2025)

$δ$-STEAL: LLM Stealing Attack with Local Differential Privacy
by: Dang, Kieu, et al.
Published: (2025)

Exploiting Novel GPT-4 APIs
by: Pelrine, Kellin, et al.
Published: (2023)

UniC-RAG: Universal Knowledge Corruption Attacks to Retrieval-Augmented Generation
by: Geng, Runpeng, et al.
Published: (2025)

Large Language Models are Advanced Anonymizers
by: Staab, Robin, et al.
Published: (2024)

Learning Obfuscations Of LLM Embedding Sequences: Stained Glass Transform
by: Roberts, Jay, et al.
Published: (2025)

Amplifying Training Data Exposure through Fine-Tuning with Pseudo-Labeled Memberships
by: Oh, Myung Gyo, et al.
Published: (2024)

Defending against Backdoor Attacks via Module Switching
by: Li, Weijun, et al.
Published: (2025)

DP-2Stage: Adapting Language Models as Differentially Private Tabular Data Generators
by: Afonja, Tejumade, et al.
Published: (2024)

Jailbreak Mimicry: Automated Discovery of Narrative-Based Jailbreaks for Large Language Models
by: Ntais, Pavlos
Published: (2025)

Unlearning at Scale: Implementing the Right to be Forgotten in Large Language Models
by: X, Abdullah
Published: (2025)

POISONCRAFT: Practical Poisoning of Retrieval-Augmented Generation for Large Language Models
by: Shao, Yangguang, et al.
Published: (2025)

Evaluating the Robustness of Large Language Model Safety Guardrails Against Adversarial Attacks
by: Young, Richard J.
Published: (2025)

Bypassing LLM Guardrails: An Empirical Analysis of Evasion Attacks against Prompt Injection and Jailbreak Detection Systems
by: Hackett, William, et al.
Published: (2025)

Siren -- Advancing Cybersecurity through Deception and Adaptive Analysis
by: Ananthanarayanan, Samhruth, et al.
Published: (2024)

Logits of API-Protected LLMs Leak Proprietary Information
by: Finlayson, Matthew, et al.
Published: (2024)

Shortcuts Arising from Contrast: Effective and Covert Clean-Label Attacks in Prompt-Based Learning
by: Xie, Xiaopeng, et al.
Published: (2024)

Scaling Patterns in Adversarial Alignment: Evidence from Multi-LLM Jailbreak Experiments
by: Nathanson, Samuel, et al.
Published: (2025)

Jailbreaking Attacks vs. Content Safety Filters: How Far Are We in the LLM Safety Arms Race?
by: Xin, Yuan, et al.
Published: (2025)

Blind Spots in the Guard: How Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems
by: Pai, Aaditya
Published: (2026)

Power-Softmax: Towards Secure LLM Inference over Encrypted Data
by: Zimerman, Itamar, et al.
Published: (2024)

Anomaly Detection in IEC-61850 GOOSE Networks: Evaluating Unsupervised and Temporal Learning for Real-Time Intrusion Detection
by: Moore, Joseph
Published: (2026)

JavelinGuard: Low-Cost Transformer Architectures for LLM Security
by: Datta, Yash, et al.
Published: (2025)

RouteScan: A Non-Intrusive Approach to Auditing MoE LLMs Safety via Expert Routing Telemetry
by: Lv, Bo, et al.
Published: (2026)

LLM Detectors Still Fall Short of Real World: Case of LLM-Generated Short News-Like Posts
by: Gameiro, Henrique Da Silva, et al.
Published: (2024)