Russell Howes
Source-listed: Researcher · Meta (FAIR)
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and PlanningSaved credit: Contributor · 2025Explore people connected to this work →
- DINOv2: Learning Robust Visual Features without SupervisionSaved credit: Contributor · 2023Explore people connected to this work →
- Demystifying CLIP DataSaved credit: Contributor · 2023Explore people connected to this work →
- The Llama 3 Herd of ModelsSaved credit: Contributor · 2024Explore people connected to this work →
- SeamlessM4T: Massively Multilingual & Multimodal Machine TranslationSaved credit: Contributor · 2023Explore people connected to this work →