Cover Image for Scaling Long Context Transformers by Bowen Yang (Cohere)
Cover Image for Scaling Long Context Transformers by Bowen Yang (Cohere)
Avatar for Studio 535
Presented by
Studio 535
All events @ 535

Scaling Long Context Transformers by Bowen Yang (Cohere)

Get Tickets
Past Event
Ticket Price
CA$5.35
Welcome! To join the event, please get your ticket below.
About Event

Bowen Yang from Cohere will give a lecture on the frontier of long contexts and transformers. The lecture will cover a range of approaches from data preparation, modeling methods, architecture innovations and framework-level optimizations, followed by Q&A.

This is an advanced lecture: Understanding of attention and ROPE is expected.

Some good reading materials:
1. Ring Attention: https://arxiv.org/abs/2310.01889
2. Rope to Nope and Back Again: https://arxiv.org/abs/2501.18795

There will be snacks and drinks.


The 535 AI Lecture Series is brought to you thanks to Cohere.

Location
Studio 535
535 Queen St E, Toronto, ON M5A 1V1, Canada
Back entrance, where the parking lot is. Door has a 535 sticker on it.
Avatar for Studio 535
Presented by
Studio 535
All events @ 535