While LLM quantization's impact on utility has been extensively explored, this work for the first time studies its adverse effects from a security perspective.

InFigure 7 of this Appendix, we show that indeed this is due to a decrease in the robustness slope. Across three different datasets, MNIST, CIFAR10, NewsGroup20, we see that increasing the number of tasks leads to a decrease in the robustness slope. Experiments on other languages For our experiments on multilingual generative models, we decided to use Greek and English because we were looking for a linguistic pair with different morphology,syntaxandphonology. This ensures that any benefits in terms of robustness are not coming from exposure to more data. Asshownin Figure 8,eventhough thetwomodels arestarting from roughly thesame perplexity,thebilingual model exhibits higher structural robustness in the presence of weight deletions.

InMTL, there areDdifferent labeled domains where data are related and the goal is to improve the predictive power of all domains altogether.

Neural Information Processing Systems http://nips.cc/

Xiaoqian Wu Shanghai Jiao Tong University enlighten@sjtu.edu.cn In Tab. 1, we conclude the notations in this work for clarity.Notation Definition r A rule. The size of the premise symbols set M . S is the symbol set, and R is the rule set. A \ B The set difference of A and B. D A very large-scale activity images database.

Human reasoning can be understood as a cooperation between the intuitive, associative "System-1" and the deliberative, logical "System-2".

Recently,there has been agrowing interest in developing saliencymethods that providevisualexplanations ofnetworkpredictions. Still,theusability ofexisting methods is limited to image classification models. To overcome this limitation, we extend the existing approaches to generategrid saliencies, which provide spatially coherent visualexplanations for(pixel-level)denseprediction networks.