Samuel McCandlish

TL;DR: GPT-3 achieves strong performance on many NLP datasets, including translation, question-answering, and cloze tasks, as well as several tasks that require on-the-fly reasoning or domain adaptation, such as unscrambling words, using a novel word in a sentence, or performing 3-digit arithmetic.

TL;DR: This article showed that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches.

TL;DR: Larger models are significantly more sample-efficient, such that optimally compute-efficient training involves training very large models on a relatively modest amount of data and stopping significantly before convergence.

TL;DR: In this article, the authors present a mathematical framework which underlies the connection between information theory and the bulk spacetime in the AdS3/CFT2 correspondence, and explain how basic geometric concepts -points, distances and angles - are reflected in kinematic space, allowing one to reconstruct a large class of spatial bulk geometries from boundary entan-gate entropies.

TL;DR: In this article, the authors present a dictionary of non-local CFT operators whose duals are simple, diffeomorphism-invariant bulk operators, such as the modular Hamiltonian, which is dual to the fluctuation in the area of a minimal surface.