Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore a critical examination of the "reasoning lie" in Large Language Models (LLMs) in this 38-minute video from Discover AI. Delve into extended reasoning concepts like CLAUSE Sonnet 3.7 and other thinking models, with insights applicable to both classical LLMs and Test-Time-Compute Scaling TTS models such as o1 and o3. Learn about the latest research findings that challenge the notion of inherent logical emergence of intelligence in AI systems. The video references work by researchers from Fudan University and Ant Group, specifically their paper "Order Doesn't Matter, But Reasoning Does: Training LLMs with Order-Centric Augmentation."
