Genre
SAFEWORLD: Geo-DiverseSafetyAlignment
Despite significant progress inthisarea, anessential factor often remains overlooked:geo-diversity. Recognizing and incorporating geographical variations [41, 40, 4, 10, 31, 6] in safety principles is crucial in the global landscape of LLM safety. Cultural norms and legal frameworks vary widely, resulting in diverse definitions of safe and acceptable behavior. As shown in Figure 1, while giving a green hatasagift might bebenign inmanycultures, itisconsidered offensiveinChina.
Symmetric Linear Bandits with Hidden Symmetry
However, it may not be available in practice.
Private Algorithms for Stochastic Saddle Points and Variational Inequalities: Beyond Euclidean Geometry
This work was done while M. Menart was at The Ohio State University.
Decision-aware learning for geographical districting
Districting is a complex combinatorial problem that consists in partitioning a geographical area into small districts.
Slice-100K: A Multimodal Dataset for Extrusion-based 3D Printing
We build our dataset from triangu-lated meshes derived from Objaverse-XL and Thingi10K datasets.
A hierarchical decomposition for explaining ML performance discrepancies
Machine learning (ML) algorithms can often differ in performance across domains.
Stabilizing Zero-Shot Prediction: A Novel Antidote to Forgetting in Continual Vision-Language Tasks Zijian Gao
Continual learning (CL) empowers pre-trained vision-language (VL) models to efficiently adapt to a sequence of downstream tasks.
