AI papers of the week - June 26th, 2024 - Consistency Models Made Easy, Instruction Pretraining, etc.

Papers covered:

Consistency Models Made Easy

Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data

Instruction Pre-Training: Language Models are Supervised Multitask Learners

Data curation via joint example selection further accelerates multimodal learning

Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?

Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMs