🧑‍🔬 James Lee-Thorp

Papers

The most cited first (OpenAlex and our library).

1. GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints (2023 · 1 012 citations)
   During autoregressive decoding, a Transformer model loads the attention keys and values from memory at every step.
2. FNet: Mixing Tokens with Fourier Transforms (2022 · 593 citations)
   The authors ask whether the self-attention sublayers of a Transformer encoder are truly necessary, or whether cheaper token-mixing operations can replace them.

Co-authors here

3 co-authors with a page, the most papers together first.

• Joshua Ainslie ×2
• Santiago Ontañón
• Sumit Sanghai

Cited by

Papers of our library cite their works 658 times (of the 140 884 papers whose bibliographies we read); 29 citing papers with a page here →.

This page in Markdown → https://md.research.elamur.ai/free/en/authors/james-lee-thorp/