
Obvious Ownership: A Sensible Humane Registry
Exploring a sensible and humane approach to ownership in engineering at SoundCloud

BanditPAM: Almost Linear-Time k-medoids Clustering via Multi-Armed Bandits
Stanford researchers have significantly sped up k-medoids clustering with their new BanditPAM algorithm, which matches sklearn.cluster.KMeans interface and can be installed with a simple pip command.

WebGPT: Improving the factual accuracy of language models through web browsing
Enhancing model accuracy by leveraging web browsing

Customizing GPT-3 for your application
Optimizing the language model GPT-3 to suit specific application needs.

Simulating matter on the quantum scale with AI
Simulating matter on the quantum scale with AI to design new materials and solve major challenges of the 21st Century

Chrome on Windows performance improvements and the journey of Native Window Occlusion
The blogpost discusses the performance improvements in Chrome on Windows by utilizing the concept of Native Window Occlusion, resulting in faster startup and fewer crashes.

Simulating matter on the quantum scale with AI
Simulating electrons in order to design new materials for clean electricity and high temperature superconductors.

Creating Interactive Agents with Imitation Learning
Creating Interactive Agents with Imitation Learning: Building a multimodal interactive agent that engages in extended and surprising physical and linguistic interactions with humans using imitation learning.

Improving language models by retrieving from trillions of tokens
Improving language models by retrieving from trillions of tokens

Language modelling at scale: Gopher, ethical considerations, and retrieval
Language modelling at scale: Gopher, ethical considerations, and retrieval

Creating Interactive Agents with Imitation Learning
Imitation learning and self-supervised learning enable the creation of a multimodal interactive agent, MIA, that successfully interacts with non-adversarial humans 75% of the time, with improved performance using hierarchical action selection.

Improving language models by retrieving from trillions of tokens
Exploring RETRO: Augmenting transformers with retrieval to improve language models