🧑‍🔬 Elie Bakouch

Affiliations

• Hugging Face · New York, United States — current · 2025–2026 on 2 papers
• Prime Intellect (United States) · San Francisco, United States — current · 2026 on 1 paper
• Princeton University · Princeton, United States — current · 2026 on 1 paper

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 1 more institution on single bylines are not listed.

Topics

Language models & pretraining1

Works by year

2020s

Papers

The most cited first (OpenAlex and our library).

1. SmolLM2: When Smol Goes Big -- Data-Centric Training of a Small Language Model (2025 · 171 citations)
   Small language models are cheap to run and fit on devices like phones. Because they have limited capacity, though, the quality and balance of their training data matter even more than for large models. In SmolLM2, the Hugging Face team treats data design as the main lever for performance. They built a 1.7B-parameter…
2. Scaling Laws and Compute-Optimal Training Beyond Fixed Training Durations (2024 · 24 citations)
   The paper questions the habit of training large language models with a cosine learning rate schedule. This schedule performs best solely when its cycle is set to equal the full training duration. As a result, the model has to be trained from scratch for every token budget, which makes scaling laws studies expensive.…

Also in our library

1 more work of theirs is in our library without a page here yet; it will appear as we publish them.

Co-authors here

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

• Loubna Ben Allal ×2
• Anton Lozhkov
• Guilherme Penedo
• Hynek Kydlíček
• Leandro Von Werra
• Lewis Tunstall
• Martin Jaggi

Cited by

Papers of our library cite their works 230 times (of the 140 884 papers whose bibliographies we read); 11 citing papers with a page here →.

This page in Markdown → https://md.research.elamur.ai/free/en/authors/elie-bakouch/