Transformers are inherently succinct (openreview.net)

Screenshot of openreview.net: Transformers are inherently succinct

First screen, captured 4 months ago.

Looks similar

Screenshot of arxiv.org: Transformers Are Inherently Succinct (2025)

Transformers Are Inherently Succinct (2025)(arxiv.org)

45 points by bearseascape 5 months ago | 6 comments

Screenshot of arxiv.org: Transformers as Support Vector Machines

Transformers as Support Vector Machines (arxiv.org)

209 points by fofoz 3 years ago | 129 comments

Screenshot of arxiv.org: Differential Transformer

Differential Transformer (arxiv.org)

525 points by weirdcat 2 years ago | 166 comments

arXiv paper introducing Diff Transformer, which computes attention as the difference of two softmax maps to cancel noise and improve long-context modeling and reduce hallucination.

differential-transformer attention-mechanism transformers large-language-models

Screenshot of arxiv.org: A Mathematical Perspective on Transformers

A Mathematical Perspective on Transformers (arxiv.org)

63 points by georgehill 2 years ago | 4 comments

Screenshot of arxiv.org: Transformer Layers as Painters

Transformer Layers as Painters (arxiv.org)

67 points by fzliu 2 years ago | 9 comments

Screenshot of cdn.openai.com: Weight-sparse transformers have interpretable circuits [pdf]

Weight-sparse transformers have interpretable circuits [pdf](cdn.openai.com)

68 points by 0x79de 11 months ago | 29 comments

Looks similar

Transformers Are Inherently Succinct (2025)(arxiv.org)

Transformers as Support Vector Machines(arxiv.org)

Differential Transformer(arxiv.org)

A Mathematical Perspective on Transformers(arxiv.org)

Transformer Layers as Painters(arxiv.org)

Weight-sparse transformers have interpretable circuits [pdf](cdn.openai.com)

Transformers as Support Vector Machines (arxiv.org)

Differential Transformer (arxiv.org)

A Mathematical Perspective on Transformers (arxiv.org)

Transformer Layers as Painters (arxiv.org)