Notes from EMNLP 2024
Last week, I was at EMNLP in Miami, and here are a few notes about what I saw at the conference.
Keynotes
The conference had three keynotes: two good and one amazing.
In the first keynote, Percy Liang talked about research on LLMs that they do at Stanford. One topic was LLM-based agents: Percy Liang predicts that LLMs are awaiting their AlphaGo moment so that we will move from coded agents; soon, the big topic will be agents trained with reinforcement learning, in other words, small black boxes operating over big black boxes. I am not completely sure about this.
The second keynote by Anca Dragan discussed reinforcement learning. As always, when talking about reinforcement learning, we saw examples of learning bad objectives that were exploited by the algorithms.