Copertina

Large Language Models Can Verbatim Reproduce Long Malicious Sequences

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Lin, Sharon, Krishnamurthy, Dvijotham, Hayes, Jamie, Shi, Chongyang, Shumailov, Ilia, Song, Shuang
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Buffer Overflow in Mixture of Experts
di: Hayes, Jamie, et al.
Pubblicazione: (2024)

Interpreting the Repeated Token Phenomenon in Large Language Models
di: Yona, Itay, et al.
Pubblicazione: (2025)

Beyond Slow Signs in High-fidelity Model Extraction
di: Foerster, Hanna, et al.
Pubblicazione: (2024)

Measuring memorization in RLHF for code completion
di: Pappu, Aneesh, et al.
Pubblicazione: (2024)

Cascading Adversarial Bias from Injection to Distillation in Language Models
di: Chaudhari, Harsh, et al.
Pubblicazione: (2025)

Stealing User Prompts from Mixture of Experts
di: Yona, Itay, et al.
Pubblicazione: (2024)

Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy
di: Hayes, Jamie, et al.
Pubblicazione: (2024)

Soft Instruction De-escalation Defense
di: Walter, Nils Philipp, et al.
Pubblicazione: (2025)

Lessons from Defending Gemini Against Indirect Prompt Injections
di: Shi, Chongyang, et al.
Pubblicazione: (2025)

Quantamination: Dynamic Quantization Leaks Your Data Across the Batch
di: Foerster, Hanna, et al.
Pubblicazione: (2026)

Demystifying Verbatim Memorization in Large Language Models
di: Huang, Jing, et al.
Pubblicazione: (2024)

Thought-Transfer: Indirect Targeted Poisoning Attacks on Chain-of-Thought Reasoning Models
di: Chaudhari, Harsh, et al.
Pubblicazione: (2026)

Locking Machine Learning Models into Hardware
di: Clifford, Eleanor, et al.
Pubblicazione: (2024)

Fairness Feedback Loops: Training on Synthetic Data Amplifies Bias
di: Wyllie, Sierra, et al.
Pubblicazione: (2024)

Achieving the Tightest Relaxation of Sigmoids for Formal Verification
di: Chevalier, Samuel, et al.
Pubblicazione: (2024)

Reasoning Introduces New Poisoning Attacks Yet Makes Them More Complicated
di: Foerster, Hanna, et al.
Pubblicazione: (2025)

Norm-Bounded Low-Rank Adaptation
di: Wang, Ruigang, et al.
Pubblicazione: (2025)

Monotone, Bi-Lipschitz, and Polyak-Lojasiewicz Networks
di: Wang, Ruigang, et al.
Pubblicazione: (2024)

SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks
di: Gao, Yue, et al.
Pubblicazione: (2023)

Machine Learning needs Better Randomness Standards: Randomised Smoothing and PRNG-based attacks
di: Dahiya, Pranav, et al.
Pubblicazione: (2023)

Keeping up with dynamic attackers: Certifying robustness to adaptive online data poisoning
di: Bose, Avinandan, et al.
Pubblicazione: (2025)

Architectural Backdoors for Within-Batch Data Stealing and Model Inference Manipulation
di: Küchler, Nicolas, et al.
Pubblicazione: (2025)

Measuring memorization in language models via probabilistic extraction
di: Hayes, Jamie, et al.
Pubblicazione: (2024)

Honeyval: A Comprehensive Evaluation Framework for LLM-powered HTTP Honeypots
di: Vero, Mark, et al.
Pubblicazione: (2026)

ceLLMate: Sandboxing Browser AI Agents
di: Meng, Luoxi, et al.
Pubblicazione: (2025)

ParaPO: Aligning Language Models to Reduce Verbatim Reproduction of Pre-training Data
di: Chen, Tong, et al.
Pubblicazione: (2025)

Beyond Labeling Oracles: What does it mean to steal ML models?
di: Shafran, Avital, et al.
Pubblicazione: (2023)

Watermarking Needs Input Repetition Masking
di: Khachaturov, David, et al.
Pubblicazione: (2025)

CrisperWhisper: Accurate Timestamps on Verbatim Speech Transcriptions
di: Wagner, Laurin, et al.
Pubblicazione: (2024)

Language Models May Verbatim Complete Text They Were Not Explicitly Trained On
di: Liu, Ken Ziyu, et al.
Pubblicazione: (2025)

The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections
di: Nasr, Milad, et al.
Pubblicazione: (2025)

Hardware and Software Platform Inference
di: Zhang, Cheng, et al.
Pubblicazione: (2024)

Revisiting Block-based Quantisation: What is Important for Sub-8-bit LLM Inference?
di: Zhang, Cheng, et al.
Pubblicazione: (2023)

UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI
di: Shumailov, Ilia, et al.
Pubblicazione: (2024)

Machine Learning Models Have a Supply Chain Problem
di: Meiklejohn, Sarah, et al.
Pubblicazione: (2025)

Confidence-aware Reward Optimization for Fine-tuning Text-to-Image Models
di: Kim, Kyuyoung, et al.
Pubblicazione: (2024)

Beyond Laplace and Gaussian: Exploring the Generalized Gaussian Mechanism for Private Machine Learning
di: Rinberg, Roy, et al.
Pubblicazione: (2025)

ImpNet: Imperceptible and blackbox-undetectable backdoors in compiled neural networks
di: Clifford, Eleanor, et al.
Pubblicazione: (2022)

When Vision Fails: Text Attacks Against ViT and OCR
di: Boucher, Nicholas, et al.
Pubblicazione: (2023)

No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms
di: Kazdan, Joshua, et al.
Pubblicazione: (2025)