Papers covered:
How Do Large Language Models Acquire Factual Knowledge During Pretraining?
Autoregressive Image Generation without Vector Quantization
Transcendence: Generative Models Can Outperform The Experts That Train Them
OpenVLA: An Open-Source Vision-Language-Action Model
DeepSeek-Coder-V2
Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMs
Large Language Models Must Be Taught to Know What They Don't Know
DataComp-LM: In search of the next generation of training sets for language models
Meta AI release news
AI papers of the week - June 19th, 2024 - Transcendence, OpenVLA, DeepSeek-Coder-V2, DataComp-LM, etc.