新闻
Surprisal Theory is Tautological (without Rational Grounding)
arXiv cs.AI · 发布于 · 阅读约3分钟
30秒读懂
- 发生了什么
- Surprisal theory, which links human reading difficulty to language model uncertainty, is logically tautological without additional constraints on the model itself.
- 为何重要
- Matters for NLP researchers building cognitive models and psycholinguists validating theories against human processing data.
- 注意
- The paper argues surprisal theory needs external constraints from cognitive theory, not just corpus-fit improvements, to make falsifiable predictions about human behavior.
收听本摘要
- language model
The Agent Architect
每周一个模式、一个权衡、一个生产事故案例。为构建智能体系统的人准备的每周简报。
每周一封邮件,一键退订。您的地址仅用于发送简报。