Thomas Scialom

Led the development of Llama 2 and Llama 3 at Meta, focusing on pre-training and post-training techniques. His research interests include reinforcement learning and multilingual language models.

Top 3 podcasts with Thomas Scialom

Ranked by the Snipd community

Jul 23, 2024 • 1h 5min

Llama 2, 3 & 4: Synthetic Data, RLHF, Agents on the path to Open Source AGI

In this engaging discussion with Thomas Scialom, a leading mind behind Llama 2 and Llama 3 at Meta, listeners dive into the fascinating world of synthetic data and reinforcement learning techniques. He reveals how Llama 3 excels with 15T tokens, leveraging primarily synthetic content for training efficiency. The importance of evaluation methods and the balance between human feedback and model training strategies takes center stage. Scialom also shares insights on the future of intelligence with advanced, multi-step agents and the evolving landscape of AI innovation.

Dec 22, 2024 • 2h 42min

S5E5 - Pourquoi le problème du DDoS est en train d'empirer

Dans ce dialogue captivant, Thomas Sialom, expert en IA chez Meta, Paul Vielle, ingénieur réseau spécialisé en reverse engineering, et Jérôme Meyer, chercheur chez Nokia Deepfield, plongent dans le monde fascinant des attaques DDoS. Ils explorent l'évolution inquiétante de ces menaces, les motivations derrière les cybercriminels, ainsi que l'impact croissant des objets connectés. Les discussions mettent également en lumière le rôle des analyses de trafic pour contrer ces attaques et l'importance d'une meilleure sécurité réseau dans un paysage technologique en constante évolution.

Sep 12, 2023 • 1h 26min

713: Llama 2, Toolformer and BLOOM: Open-Source LLMs with Meta's Dr. Thomas Scialom

Dr. Thomas Scialom discusses Llama 2, Toolformer, and BLOOM: open-source LLMs. Topics include AGI, RLHF in AI, and advice for AI entrepreneurs. Exploring Toolformer's capabilities, the Galactica project, and AI models' responsible use. Insights on developing large-scale AI projects and the future of AI industry.