Record
- Authors
- Merrill, N.; zekem
- Year
- 2026
- Title
- Defeating Introspection Adapters (and Why Threat Models Matter)
- Type
- Web page
- Links
- https://www.lesswrong.com/posts/mhFqC9B5EpaxHNdt6/defeating-introspection-adapters-and-why-threat-models
- Our note
- none yet
- Verification
- url-ok: the web address responds (the page itself is not checked).
https://www.lesswrong.com/posts/mhFqC9B5EpaxHNdt6/defeating-introspection-adapters-and-why-threat-models: HTTP 200, checked 2026-11-14 - Id
merrill-2026-defeating-introspection-adapters-why-threat-models
Cited on
- Introspection Adapters: Training LLMs to Report Their Learned Behaviors read status: not stated
As cited
The pages cite this work in one form.
-
Merrill, N., and zekem. (2026, June 4). Defeating Introspection Adapters (and Why Threat Models Matter). LessWrong. https://www.lesswrong.com/posts/mhFqC9B5EpaxHNdt6/defeating-introspection-adapters-and-why-threat-models
BibLaTeX
@online{merrill-2026-defeating-introspection-adapters-why-threat-models,
author = {Merrill, N. and {zekem}},
title = {{Defeating Introspection Adapters (and Why Threat Models Matter)}},
date = {2026},
url = {https://www.lesswrong.com/posts/mhFqC9B5EpaxHNdt6/defeating-introspection-adapters-and-why-threat-models}
}