Copertina

Virus: Harmful Fine-tuning Attack for Large Language Models Bypassing Guardrail Moderation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Huang, Tiansheng, Hu, Sihao, Ilhan, Fatih, Tekin, Selim Furkan, Liu, Ling
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Cryptography and Security Artificial Intelligence Computation and Language Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
di: Huang, Tiansheng, et al.
Pubblicazione: (2024)

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation
di: Huang, Tiansheng, et al.
Pubblicazione: (2024)

Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack
di: Huang, Tiansheng, et al.
Pubblicazione: (2024)

Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable
di: Huang, Tiansheng, et al.
Pubblicazione: (2025)

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning
di: Huang, Tiansheng, et al.
Pubblicazione: (2024)

MELT: A Behavioral Trace Dataset for High-Risk Memecoin Launch Detection
di: Hu, Sihao, et al.
Pubblicazione: (2026)

CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning
di: Yi, Biao, et al.
Pubblicazione: (2025)

LLM-TOPLA: Efficient LLM Ensemble by Maximising Diversity
di: Tekin, Selim Furkan, et al.
Pubblicazione: (2024)

Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack
di: Huang, Tiansheng, et al.
Pubblicazione: (2024)

Pharmacist: Safety Alignment Data Curation for Large Language Models against Harmful Fine-tuning
di: Liu, Guozhi, et al.
Pubblicazione: (2025)

OpenGuardrails: A Configurable, Unified, and Scalable Guardrails Platform for Large Language Models
di: Wang, Thomas, et al.
Pubblicazione: (2025)

EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language Models
di: Wu, Jialin, et al.
Pubblicazione: (2025)

Window-based Membership Inference Attacks Against Fine-tuned Large Language Models
di: Chen, Yuetian, et al.
Pubblicazione: (2026)

Why LLM Safety Guardrails Collapse After Fine-tuning: A Similarity Analysis Between Alignment and Fine-tuning Datasets
di: Hsiung, Lei, et al.
Pubblicazione: (2025)

H3Fusion: Helpful, Harmless, Honest Fusion of Aligned LLMs
di: Tekin, Selim Furkan, et al.
Pubblicazione: (2024)

Black-Box Guardrail Reverse-engineering Attack
di: Yao, Hongwei, et al.
Pubblicazione: (2025)

Practical Membership Inference Attacks against Fine-tuned Large Language Models via Self-prompt Calibration
di: Fu, Wenjie, et al.
Pubblicazione: (2023)

Mitigating Fine-tuning based Jailbreak Attack with Backdoor Enhanced Safety Alignment
di: Wang, Jiongxiao, et al.
Pubblicazione: (2024)

Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models
di: Yi, Biao, et al.
Pubblicazione: (2025)

Safety Anchor: Defending Harmful Fine-tuning via Geometric Bottlenecks
di: Lu, Guoxin, et al.
Pubblicazione: (2026)

LLM-Virus: Evolutionary Jailbreak Attack on Large Language Models
di: Yu, Miao, et al.
Pubblicazione: (2024)

No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks
di: Leong, Chak Tou, et al.
Pubblicazione: (2024)

"Moralized" Multi-Step Jailbreak Prompts: Black-Box Testing of Guardrails in Large Language Models for Verbal Attacks
di: Wang, Libo
Pubblicazione: (2024)

Jailbreaking Large Language Models Against Moderation Guardrails via Cipher Characters
di: Jin, Haibo, et al.
Pubblicazione: (2024)

ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models
di: Zhao, Yunhan, et al.
Pubblicazione: (2026)

SGuard-v1: Safety Guardrail for Large Language Models
di: Lee, JoonHo, et al.
Pubblicazione: (2025)

Private Fine-tuning of Large Language Models with Zeroth-order Optimization
di: Tang, Xinyu, et al.
Pubblicazione: (2024)

Auto-Tuning Safety Guardrails for Black-Box Large Language Models
di: Abdulkadir, Perry
Pubblicazione: (2025)

Privately Learning from Graphs with Applications in Fine-tuning Large Language Models
di: Yin, Haoteng, et al.
Pubblicazione: (2024)

LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models
di: Ran, Delong, et al.
Pubblicazione: (2025)

BitBypass: A New Direction in Jailbreaking Aligned Large Language Models with Bitstream Camouflage
di: Nakka, Kalyan, et al.
Pubblicazione: (2025)

Privacy in Large Language Models: Attacks, Defenses and Future Directions
di: Li, Haoran, et al.
Pubblicazione: (2023)

ADVERSA: Measuring Multi-Turn Guardrail Degradation and Judge Reliability in Large Language Models
di: Owiredu-Ashley, Harry
Pubblicazione: (2026)

Triaging Threats to Specialized Guardrails
di: Mo, Wenjie Jacky, et al.
Pubblicazione: (2026)

In AI Sweet Harmony: Sociopragmatic Guardrail Bypasses and Evaluation-Awareness in OpenAI gpt-oss-20b
di: Durner, Nils
Pubblicazione: (2025)

Fun-tuning: Characterizing the Vulnerability of Proprietary LLMs to Optimization-based Prompt Injection Attacks via the Fine-Tuning Interface
di: Labunets, Andrey, et al.
Pubblicazione: (2025)

Prompt Stealing Attacks Against Large Language Models
di: Sha, Zeyang, et al.
Pubblicazione: (2024)

SCOUT: A Defense Against Data Poisoning Attacks in Fine-Tuned Language Models
di: Afane, Mohamed, et al.
Pubblicazione: (2025)

Privacy-Preserving Parameter-Efficient Fine-Tuning for Large Language Model Services
di: Li, Yansong, et al.
Pubblicazione: (2023)

Large Language Models are Good Attackers: Efficient and Stealthy Textual Backdoor Attacks
di: Li, Ziqiang, et al.
Pubblicazione: (2024)