🎙️ w/ @MelMitchell1 on testing LLMs and the debate over understanding.
Why …
🎙️ w/ @MelMitchell1 on testing LLMs and the debate over understanding.
Why LLMs should be probed differently than humans, OpenAI’s lack of transparency, the human drive for compression, calling for a new science of intelligence, and much more.
Currents 088: Melanie Mitchell on AI Measurement and Understanding – The Jim Rutt Show
Jim talks with Melanie Mitchell about her critique of applying standardized exams to LLMs and the debate over understanding in AI. They discuss ChatGPT and GPT-4’s performance on standardized exams, questioning the underlying assumptions, OpenAI’s lack of transparency, soon-to-be-released open-source LLMs, prompt engineering, making GPT its own skyhook to reduce hallucinations, the number of parameters in …
Tagged with :