The Implications of Linguistic Illegibility for LLM Security

发生了什么

LLMs are trained to generate natural language. However, various strands of evidence indicate that an LLM's externalized linguistic outputs and mechanistically-extracted linguistic features can be an unreliable lens for understanding internal model computation. We introduce the term ``linguistic illegibility'' to broadly refer to scenarios in which an LLM's externalized or mechanistically-probed language artifacts fa…

摘要按规则整理自下方来源原文

时间线

1. 19:00
   Hacker News 最先出现HHacker News
2. 05:34
   Lobsters 出现讨论LLobsters

来源