Marcus 点出了LLM的核心短板——缺乏显式世界模型，做AI推理和知识表示的开发者值得关注，看完会重新思考LLM的局限性。
Gary Marcus 在推文中指出，世界模型（world model）并非新概念，已在象棋程序、导航系统、维基百科等系统中存在多年，它们是对对象、地点、事件、机制等可推理内容的显式表示。然而，当前的大语言模型（LLM）缺乏这种显式世界模型。Marcus 强调，大多数世界模型是手工构建的，真正的挑战在于如何从数据中自动获取它们。这引发了关于AI系统如何更好地理解和推理世界的讨论。
World models have existed for years (though not in LLMs); I take them to be explicit representation ...
World models have existed for years (though not in LLMs); I take them to be explicit representation of objects, places, events, mechanisms etc you can reason over. Chess computers have them (board, pieces, moves, history) Nav systems have them (roads, times, etc) Wikipedia has one of a sort (when people were born, where they died, etc) They often work great (though again LLMs lack them). But most are hand-engineered. The trick is to acquire them from data. Fer Cuadra 🇦🇷🏆 @fernandoquadra @GaryMarcus Sorry @GaryMarcus but what exactly is a "world model"? He talks about it as if it already exists... is there actually a "world model" present here? 🔗 View Quoted Tweet 💬 3 🔄 3 ❤️ 12 👀 2050 📊 4 ⚡