Talks

Unveiling the reasoning behaviour of medical large language models — from the paper Critique of Impure Reason.

Accelerating CPU-based fine-tuning for LLMs to approach GPU-level performance for medical summarization.