英文幻灯片，中文演讲。
演讲预告：https://mp.weixin.qq.com/s/DBf4AVto9oIvhgYpCHG3Dw
演讲幻灯片：https://ysymyth.github.io/papers/from_language_models_to_language_agents.pdf
个人主页：https://ysymyth.github.io/
Abstract：
Large language models (LLM) have revolutionized natural language processing, but they are trained to write, not to act or make decisions. We introduce several ideas (ReAct, Reflexion, Tree of Thoughts) to transform them into language agents that autonomously interact with the world, and several new benchmarks (WebShop, InterCode, Collie) to develop and evaluate such agents without relying on human preference labeling or LLM scoring.
