harrymayne

Karma: 37

A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior

harrymayne, Justin kang, Dewi Gould and noahys

26 Feb 2026 17:03 UTC

23 points

0 comments4 min readLW link

LLMs Don’t Know Their Own Decision Boundaries. Why Is This Important?

harrymayne and ryanothnielkearns

17 Sep 2025 16:39 UTC

9 points

0 comments5 min readLW link

(arxiv.org)

Are recent LLMs better at reasoning or better at memorizing?

Jude Khouja, harrymayne, ryanothnielkearns and karolinakorgul

7 Mar 2025 2:44 UTC

11 points

0 comments4 min readLW link