13 citations · 27 across the 9 of their papers we have counts for
7 papers · 1 filter
From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers
Jingtong Su, Julia Kempe, Karen Ullrich
Transformers have achieved state-of-the-art performance across language and vision tasks. This success drives the imperative to interpret their internal mechanisms with the dual go…
Mission Impossible: A Statistical Perspective on Jailbreaking LLMs
Jingtong Su, Julia Kempe, Karen Ullrich
Large language models (LLMs) are trained on a deluge of text data with limited quality control. As a result, LLMs can exhibit unintended or even harmful behaviours, such as leaking…
On the Robustness of Neural Collapse and the Neural Collapse of Robustness
Jingtong Su, Ya Shi Zhang, Nikolaos Tsilivis +1
Neural Collapse refers to the curious phenomenon in the end of training of a neural network, where feature vectors and classification weights converge to a very simple geometrical…
Wavelets Beat Monkeys at Adversarial Robustness
Jingtong Su, Julia Kempe
Research on improving the robustness of neural networks to adversarial noise - imperceptible malicious perturbations of the data - has received significant attention. The currently…
On the Connection between Invariant Learning and Adversarial Training for Out-of-Distribution Generalization
Shiji Xin, Yifei Wang, Jingtong Su +1
Despite impressive success in many tasks, deep learning models are shown to rely on spurious features, which will catastrophically fail when generalized to out-of-distribution (OOD…
Can we achieve robustness from data alone?
Nikolaos Tsilivis, Jingtong Su, Julia Kempe
We introduce a meta-learning algorithm for adversarially robust classification. The proposed method tries to be as model agnostic as possible and optimizes a dataset prior to its d…
Sanity-Checking Pruning Methods: Random Tickets can Win the Jackpot
Jingtong Su, Yihang Chen, Tianle Cai +4
Network pruning is a method for reducing test-time computational resource requirements with minimal performance degradation. Conventional wisdom of pruning algorithms suggests that…