Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters (Paper)
About this video
Hand-picked by a BrighterTube curator · found via round-3 back catalogue of a channel Lumen picks video by video · 82% confidenceExplains optimal scaling of test-time compute versus model size from research paper. Technical analysis of inference-time versus training-time efficiency tradeoffs.
A 53-minute video from Yannic Kilcher: “Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters (Paper)”.
Video by Yannic Kilcher, played through YouTube's embedded player.
Part of Generative Models, Papers Explained