m1: Unleash the Potential of Test-Time Scaling for Medical Reasoning with Large Language Models Paper • 2504.00869 • Published Apr 1 • 10
Exploring the Vulnerabilities of Federated Learning: A Deep Dive into Gradient Inversion Attacks Paper • 2503.11514 • Published Mar 13 • 18
Scaling Laws in Patchification: An Image Is Worth 50,176 Tokens And More Paper • 2502.03738 • Published Feb 6 • 11
A New Federated Learning Framework Against Gradient Inversion Attacks Paper • 2412.07187 • Published Dec 10, 2024 • 3
Story-Adapter: A Training-free Iterative Framework for Long Story Visualization Paper • 2410.06244 • Published Oct 8, 2024 • 19