mailitics

Tag: been

  • Why We’ve Been Optimizing the Wrong Thing in LLMs for Years

    Why We’ve Been Optimizing the Wrong Thing in LLMs for Years The simple shift in training that unlocks foresight, faster inference, and better reasoning. The post Why We’ve Been Optimizing the Wrong Thing in LLMs for Years appeared first on Towards Data Science. Moulik Gupta Go to original source

  • Oscillatory Coordination in Cognitive Architectures: Old Dog, New Math

    Oscillatory Coordination in Cognitive Architectures: Old Dog, New Math Been working in AI since before it was cool (think 80s expert systems, not ChatGPT hype). Lately I’ve been developing this cognitive architecture called OGI that uses Top-K gating between specialized modules. Works well, proved the stability, got the complexity down to O(k²). But something’s been…