Simon Willison reveals a counterintuitive fact: ChatGPT's voice mode runs on an older, weaker GPT-4o-era model, creating a massive gap between user expectations and reality.
Reward hacking presents challenges in reinforcement learning due to flaws in reward functions, particularly impacting language models, necessitating further research and mitigation strategies.
Lilian Weng's new article deeply explores the evolution and new features of Transformers, revealing their ongoing impact in natural language processing.