7 papers
Revealing Epistemic Uncertainty in MLLMs via Causal-Invariant Masking
Haoyang Luo, Linwei Tao, Jie Gui +4
Multimodal Large Language Models (MLLMs) suffer from hallucinations, creating a critical need for Uncertainty Quantification (UQ) to ensure reliable deployment. However, existing a…
DEFUSE: Generalizable Backdoor Defense for Self-Supervised Encoders with Generative Priors
Tuo Chen, Jie Gui, Minjing Dong +4
Self-supervised learning (SSL) encoders are vulnerable to backdoor attacks, posing threats to both visual SSL encoders and vision-language encoders. Existing defenses are typically…
Robust Bimanual Vision-Language-Action Models via Embarrassingly Simple Modality Masking
Dongzhou Cheng, Ziang Li, Yixiao Zhou +6
Query-based Vision-Language-Action (VLA) models offer low-latency inference that is attractive for bimanual robotic manipulation, but we observe that they can still exhibit discont…
The Platonic Defense: Backdoor Defense for Self-Supervised Encoders in the Era of Large Scale Pre-training
Tuo Chen, Minjing Dong, Benlei Cui +2
Self-supervised learning (SSL) pretrained models have become a dominant paradigm for visual representation learning, but they are vulnerable to backdoor attacks. Existing defenses…
Dropping Experts, Recombining Neurons: Retraining-Free Pruning for Sparse Mixture-of-Experts LLMs
Yixiao Zhou, Ziyu Zhao, Dongzhou Cheng +6
Sparse Mixture-of-Experts (SMoE) architectures are widely used in large language models (LLMs) due to their computational efficiency. However, though only a few experts are activat…
On the Evolution of Federated Post-Training Large Language Models: A Model Accessibility View
Tao Guo, Junxiao Wang, Fushuo Huo +4
Federated Learning (FL) enables training models across decentralized data silos while preserving client data privacy. Recent research has explored efficient methods for post-traini…
Backdooring Self-Supervised Contrastive Learning by Noisy Alignment
Tuo Chen, Jie Gui, Minjing Dong +3
Self-supervised contrastive learning (CL) effectively learns transferable representations from unlabeled data containing images or image-text pairs but suffers vulnerability to dat…