Pith. sign in

Transformers learn in-context by gradient descent

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

citation-role summary

other 1

citation-polarity summary

fields

cs.LG 1

years

2025 1

verdicts

CONDITIONAL 1

roles

other 1

polarities

unclear 1

representative citing papers

Next-token pretraining implies in-context learning

cs.LG · 2025-05-23 · conditional · novelty 7.0

A well-trained next-token predictor's in-context loss equals the conditional entropy of the data process, which must decrease with context for stationary data.

citing papers explorer

Showing 1 of 1 citing paper.

  • Next-token pretraining implies in-context learning cs.LG · 2025-05-23 · conditional · none · ref 6

    A well-trained next-token predictor's in-context loss equals the conditional entropy of the data process, which must decrease with context for stationary data.