Defending Against Malicious Finetuning by Scaling Train-time Adversarial Attacks
cs.CL, cs.AI
Submitted: 2026-06-06
Updated: 2026-09-27
Code: https://github.com/haomingwen/patcher
Terminology
Sources
- Sharpness-Aware Minimization for Efficiently Improving Generalization
- Explaining and Harnessing Adversarial Examples
- The Llama 3 Herd of Models
- Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning
- Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
- Decoupled Weight Decay Regularization
- Towards Deep Learning Models Resistant to Adversarial Attacks
- Fine-tuning can cripple your foundation model; preserving features may be the solution
- AntiDote: Bi-level Adversarial Training for Tamper-Resistant LLMs
- Qwen3 Technical Report
- Qwen2 Technical Report
- Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models
- Universal and Transferable Adversarial Attacks on Aligned Language Models
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering