ArXiv TLDR

Jiaming Zhang

3 papers · Latest:

Machine Learning

Augmented Lagrangian Multiplier Network for State-wise Safety in Reinforcement Learning

ALaM introduces an augmented Lagrangian multiplier network for stable learning of state-wise safety constraints in reinforcement learning, outperforming baselines.

2605.00667
Machine Learning

Benign Overfitting in Adversarial Training for Vision Transformers

This paper provides the first theoretical analysis showing benign overfitting in adversarially trained Vision Transformers, leading to strong generalization.

2604.19724
Cryptography & Security

ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety

ProjLens demystifies MLLM backdoors, revealing how projector fine-tuning introduces vulnerabilities and detailing their low-rank structure and activation mechanism.

2604.19083

📬 Weekly AI Paper Digest

Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.