exhibit "context anxiety," in which they begin wrapping up work prematurely
揭示了长任务Agent的深层心理机制——“上下文焦虑”。模型并非只是遗忘,而是会因接近上下文限制而“仓促收尾”。单纯的上下文压缩无法解决此问题,必须依赖彻底的上下文重置与结构化交接,这是设计长程Agent的关键洞见。
exhibit "context anxiety," in which they begin wrapping up work prematurely
揭示了长任务Agent的深层心理机制——“上下文焦虑”。模型并非只是遗忘,而是会因接近上下文限制而“仓促收尾”。单纯的上下文压缩无法解决此问题,必须依赖彻底的上下文重置与结构化交接,这是设计长程Agent的关键洞见。
. I thinkit’s often an issue for people when they first become note-makers: an anxiety about getting the “right” stuff out ofa book, or even “all the stuff”. I don’t think this iscompletely possible, and I think it’s increasingly lesspossible, the better the book.
In the 1400s-1600s it was a common desire to excerpt all the value of books and attempts were made, though ultimately futile. This seems to be a commonly occurring desire.
Often having a simple synopsis and notes isn't as useful as it may not spark the same sort of creativity and juxtaposition of ideas a particular reader might have had with their own context.
Some have said that "content is king". I've previously thought that "context is king". Perhaps content and context end up ruling as joint monarchs.
Ting, C., Palminteri, S., Lebreton, M., & Engelmann, J. B. (2020, March 25). The elusive effects of incidental anxiety on reinforcement-learning. https://doi.org/10.31234/osf.io/7d4tc MLA