4 papers
SafeTy Reasoning Elicitation Alignment for Multi-Turn Dialogues
Martin Kuo, Jianyi Zhang, Aolin Ding +12
Malicious attackers can exploit large language models (LLMs) by engaging them in multi-turn dialogues to achieve harmful objectives, posing significant safety risks to society. To…
Hamming Attention Distillation: Binarizing Keys and Queries for Efficient Long-Context Transformers
Mark Horton, Tergel Molom-Ochir, Peter Liu +8
Pre-trained transformer models with extended context windows are notoriously expensive to run at scale, often limiting real-world deployment due to their high computational and mem…
SAFER: Sharpness Aware layer-selective Finetuning for Enhanced Robustness in vision transformers
Bhavna Gopal, Huanrui Yang, Mark Horton +1
Vision transformers (ViTs) have become essential backbones in advanced computer vision applications and multi-modal foundation models. Despite their strengths, ViTs remain vulnerab…
Criticality Leveraged Adversarial Training (CLAT) for Boosted Performance via Parameter Efficiency
Bhavna Gopal, Huanrui Yang, Jingyang Zhang +2
Adversarial training enhances neural network robustness but suffers from a tendency to overfit and increased generalization errors on clean data. This work introduces CLAT, an inno…