Loading profile…

Loading activity…

World & Business

Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

sway (@swaystar123) · Digg

sway

@swaystar123

Vegan btw, working on diffusion models 日本語できる, 中文学习中 Building http://languagespeedrun.com, working at @LumaLabsAI

following
followers

Featured in

AI

LLM generalization may swing during pre-training despite stable training loss

askalphaxiv shares a paper describing “mode-hopping,” in which LLMs repeatedly shift between shallow pattern-matching and generalization.

6 Sources, 18d ago, updated 18d ago

Connections

@SwayStar123