Copertina

Self-Mined Hardness for Safety Fine-Tuning

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Gupta, Prakhar, Shah, Garv, Zhang, Donghua
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning Artificial Intelligence Cryptography and Security
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Safe Delta: Consistently Preserving Safety when Fine-Tuning LLMs on Diverse Datasets
di: Lu, Ning, et al.
Pubblicazione: (2025)

When Safety Geometry Collapses: Fine-Tuning Vulnerabilities in Agentic Guard Models
di: Hossain, Ismail, et al.
Pubblicazione: (2026)

SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance-Diversity Data Selection
di: Chen, Shuhao, et al.
Pubblicazione: (2026)

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

The Data Minimization Principle in Machine Learning
di: Ganesh, Prakhar, et al.
Pubblicazione: (2024)

PrivTune: Efficient and Privacy-Preserving Fine-Tuning of Large Language Models via Device-Cloud Collaboration
di: Liu, Yi, et al.
Pubblicazione: (2025)

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

Secure LLM Fine-Tuning via Safety-Aware Probing
di: Wu, Chengcan, et al.
Pubblicazione: (2025)

PACZero: PAC-Private Fine-Tuning of Language Models via Sign Quantization
di: Ertan, Murat Bilgehan, et al.
Pubblicazione: (2026)

Acceptance Cards:A Four-Diagnostic Standard for Safe Fine-Tuning Defense Claims
di: Konrad, Phongsakon Mark, et al.
Pubblicazione: (2026)

Semantic-Aware Contrastive Fine-Tuning: Boosting Multimodal Malware Classification with Discriminative Embeddings
di: Sanchez, Ivan Montoya, et al.
Pubblicazione: (2025)

CTIGuardian: A Few-Shot Framework for Mitigating Privacy Leakage in Fine-Tuned LLMs
di: Arachchige, Shashie Dilhara Batan, et al.
Pubblicazione: (2025)

Silent Sabotage During Fine-Tuning: Few-Shot Rationale Poisoning of Compact Medical LLMs
di: Xie, Jingyuan, et al.
Pubblicazione: (2026)

Security in the Fine-Tuning Lifecycle of Large Language Models: Threats, Defenses,Evaluation, and Future Directions
di: Li, Wenjuan, et al.
Pubblicazione: (2026)

DP-Adam-AC: Privacy-preserving Fine-Tuning of Localizable Language Models Using Adam Optimization with Adaptive Clipping
di: Yang, Ruoxing
Pubblicazione: (2025)

Enhancing Continual Learning for Software Vulnerability Prediction: Addressing Catastrophic Forgetting via Hybrid-Confidence-Aware Selective Replay for Temporal LLM Fine-Tuning
di: Dou, Xuhui, et al.
Pubblicazione: (2026)

AlphaAlign: Incentivizing Safety Alignment with Extremely Simplified Reinforcement Learning
di: Zhang, Yi, et al.
Pubblicazione: (2025)

Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs
di: Jiang, Yukun, et al.
Pubblicazione: (2026)

Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense
di: Min, Rui, et al.
Pubblicazione: (2024)

CyberCScope: Mining Skewed Tensor Streams and Online Anomaly Detection in Cybersecurity Systems
di: Nakamura, Kota, et al.
Pubblicazione: (2025)

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

Double-I Watermark: Protecting Model Copyright for LLM Fine-tuning
di: Li, Shen, et al.
Pubblicazione: (2024)

Self-interpreting Adversarial Images
di: Zhang, Tingwei, et al.
Pubblicazione: (2024)

The Janus Interface: How Fine-Tuning in Large Language Models Amplifies the Privacy Risks
di: Chen, Xiaoyi, et al.
Pubblicazione: (2023)

Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks
di: Poppi, Samuele, et al.
Pubblicazione: (2024)

GraphIP-Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?
di: Zhao, Kaixiang, et al.
Pubblicazione: (2026)

Exploring Query Efficient Data Generation towards Data-free Model Stealing in Hard Label Setting
di: Pei, Gaozheng, et al.
Pubblicazione: (2024)

Saffron-1: Safety Inference Scaling
di: Qiu, Ruizhong, et al.
Pubblicazione: (2025)

Quantized Delta Weight Is Safety Keeper
di: Liu, Yule, et al.
Pubblicazione: (2024)

SafeMLRM: Demystifying Safety in Multi-modal Large Reasoning Models
di: Fang, Junfeng, et al.
Pubblicazione: (2025)

UpSafe$^\circ$C: Upcycling for Controllable Safety in Large Language Models
di: Sun, Yuhao, et al.
Pubblicazione: (2025)

I can't see it but I can Fine-tune it: On Encrypted Fine-tuning of Transformers using Fully Homomorphic Encryption
di: Panzade, Prajwal, et al.
Pubblicazione: (2024)

Fine-Tuning Language Models with Differential Privacy through Adaptive Noise Allocation
di: Li, Xianzhi, et al.
Pubblicazione: (2024)

Checkpoint-GCG: Auditing and Attacking Fine-Tuning-Based Prompt Injection Defenses
di: Yang, Xiaoxue, et al.
Pubblicazione: (2025)

Pruning for Protection: Increasing Jailbreak Resistance in Aligned LLMs Without Fine-Tuning
di: Hasan, Adib, et al.
Pubblicazione: (2024)

Why Safety Probes Catch Liars But Miss Fanatics
di: Haralambiev, Kristiyan
Pubblicazione: (2026)

Understanding the Effects of Safety Unalignment on Large Language Models
di: Halloran, John T.
Pubblicazione: (2026)

Uncovering Logit Suppression Vulnerabilities in LLM Safety Alignment
di: Li, Yuxi, et al.
Pubblicazione: (2024)

ObfuscaTune: Obfuscated Offsite Fine-tuning and Inference of Proprietary LLMs on Private Datasets
di: Frikha, Ahmed, et al.
Pubblicazione: (2024)

LMEraser: Large Model Unlearning through Adaptive Prompt Tuning
di: Xu, Jie, et al.
Pubblicazione: (2024)