Michael Backes
~Michael_Backes3
18
论文总数
9.0
年均投稿
平均评分
接收情况9/18
会议分布
ICLR
12
NeurIPS
4
ICML
2
发表论文 (18 篇)
20258 篇
4
SaLoRA: Safety-Alignment Preserved Low-Rank Adaptation
ICLR 2025Poster
4
Finding and Reactivating Post-Trained LLMs' Hidden Safety Mechanisms
NeurIPS 2025Poster
4
Adjacent Words, Divergent Intents: Jailbreaking Large Language Models via Task Concurrency
NeurIPS 2025Poster
3
Provably Cost-Sensitive Adversarial Defense via Randomized Smoothing
ICML 2025Poster
4
ACE: Attack Combo Enhancement Against Machine Learning Models
ICLR 2025withdrawn
4
POST: A Framework for Privacy of Soft-prompt Transfer
ICLR 2025Rejected
4
Efficient and Privacy-Preserving Soft Prompt Transfer for LLMs
ICML 2025Poster
3
Captured by Captions: On Memorization and its Mitigation in CLIP Models
ICLR 2025Poster
202410 篇
5
Transferable Availability Poisoning Attacks
ICLR 2024Rejected
5
Generating Less Certain Adversarial Examples Improves Robust Generalization
ICLR 2024withdrawn
4
Provably Robust Cost-Sensitive Learning via Randomized Smoothing
ICLR 2024withdrawn
4
Prompt Backdoors in Visual Prompt Learning
ICLR 2024withdrawn
4
Localizing Memorization in SSL Vision Encoders
NeurIPS 2024Poster
4
Last One Standing: A Comparative Analysis of Security and Privacy of Soft Prompt Tuning, LoRA, and In-Context Learning
ICLR 2024Rejected
4
Composite Backdoor Attacks Against Large Language Models
ICLR 2024withdrawn
4
Robustness Over Time: Understanding Adversarial Examples’ Effectiveness on Longitudinal Versions of Large Language Models
ICLR 2024Rejected
4
Memorization in Self-Supervised Learning Improves Downstream Generalization
ICLR 2024Poster
4
Open LLMs are Necessary for Current Private Adaptations and Outperform their Closed Alternatives
NeurIPS 2024Poster