Live from the web

Public annotations

Clips, highlights, and commentary from people browsing together. Open one to follow its author or join the reply thread in the Basement sidebar.

@peroni813 followers
arxiv.org/abs/2307.03172
“While recent language models have the ability to take long contexts as input, relatively little is known about how well they use longer context. We analyze the performance of language models on two tasks that require identifying relevant information in their input contexts: multi”

Such an interesting paper on how fresh context loops are more effecient.

View public artifact
@peroni813 followers
arxiv.org/abs/2307.03172
“While recent language models have the ability to take long contexts as input, relatively little is known about how well they use longer context. We analyze the performance of language models on two tasks that require identifying relevant information in their input contexts: multi”

Such an interesting paper on how fresh context loops are more effecient.

View public artifact