thinker

active

thinker:johannes-von-oswald

Johannes von Oswald

Transformers learn in-context by gradient descent.

Authored

0

Introduces

1

Studies

0

Affiliations

0

Cited by

0

More papers — OpenAlex / S2

Originates (1)

finding

Transformers learn in-context by gradient descent, functioning as mesa-optimizers that learn internal models in real time

Other inbound relations (4)

authoredvon Oswald et al. (2023) Transformers learn in-context by gradient descent(artifact)
citesSemantic Anchoring in LLMs: Thresholds, Transfer, and Geometric Correlates(artifact)
citesWhy Learning Requires Feeling(paper)
mentionsSemantic Anchoring in LLMs: Thresholds, Transfer, and Geometric Correlates(artifact)

Recent mentions (4)