Sveriges mest populära poddar

Abstracts: NeurIPS 2024 with Weizhu Chen 

8 min•6 december 2024

Om avsnittet

Next-token prediction trains a language model on all tokens in a sequence. VP Weizhu Chen discusses his team’s 2024 NeurIPS paper on how distinguishing between useful and “noisy” tokens in pretraining can improve token efficiency and model performance.

Read the paper

Get the code

Fler avsnitt

Microsoft Research Podcast med Researchers across the Microsoft research community finns tillgänglig på flera plattformar. Informationen på denna sida kommer från offentliga podd-flöden.

Abstracts: NeurIPS 2024 with Weizhu Chen

Om avsnittet

Fler avsnitt

Learn the rules: The hacker mindset

Microsoft Research Focus – Edition #97

Called to serve: Tech, research, and positive impact with Chris White

What AI gets wrong and what failure teaches us

Can we AI our way to a more sustainable world?

Ideas: Steering AI toward the work future we want

Will machines ever be intelligent?

Trailer: The Shape of Things to Come

Ideas: Community building, machine learning, and the future of AI

Ideas: More AI-resilient biosecurity with the Paraphrase Project

Abstracts: NeurIPS 2024 with Weizhu Chen 