A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Zheng Liang, Hao He, Ceyuan Yang, et al.
This paper establishes scaling laws for Diffusion Transformers (DiT) across compute budgets from 1e17 to 6e18 FLOPs, confirming power-law relationships that guide efficient training and model design.
Michalis Faloutsos, Petros Faloutsos, Christos Faloutsos
This paper discovers simple power-laws governing the Internet topology, showing high correlation with real data despite network growth.
Nischay Dhankhar, Dos Baha, Abulhair Saparov
This paper establishes scaling laws for hypernetwork-based knowledge injection into LLMs, showing power-law scaling and superior OOD generalization.