Deep Learning
TabularBench: Benchmarking Adversarial Robustness for Tabular Deep Learning in Real-world Use-cases
We evaluated adversarial robustness with CAA, an ensemble of gradient and search attacks which was recently demonstrated as the most effective attack against a tabular model.
One-to-Normal: Anomaly Personalization for Few-shot Anomaly Detection
Traditional Anomaly Detection (AD) methods have predominantly relied on unsupervised learning from extensive normal data.
Understanding Transformer Reasoning Capabilities via Graph Algorithms Clayton Sanford
Which transformer scaling regimes are able to perfectly solve different classes of algorithmic problems?
V-InFoR: A Robust Graph Neural Networks Explainer for Structurally Corrupted Graphs
Equally contribute to this research.
Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning Shuguang Y u
Before deploying any newly developed policy, it is important to assess its impact. In many high-stakes domains, it is risky or unethical to implement such policies directly for online evaluation.
8eb88844dafefa92a26aaec9f3acad93-Paper-Datasets_and_Benchmarks_Track.pdf
Ideally,languagemodelswould reflect the cultural norms of various regions around the world and generate culturally appropriate content when responding inlocallanguages oftheregions, unless otherwise specified.
