Research agenda: Interpretive debate

Shi18 Jun 2026 23:46 UTC

18 points

0 comments7 min readLW link

Does it feel any different to be reverse-chiral life?

jessicata18 Jun 2026 22:56 UTC

9 points

0 comments10 min readLW link

Reinforcement learning towards broadly and persistently beneficial models

papetoast18 Jun 2026 22:11 UTC

11 points

0 comments1 min readLW link

(alignment.openai.com)

The distillation double bind: Distilling misaligned models either transfers misalignment or it doesn’t

Alek Westover, SebastianP, Alexa Pan and Jozdien

18 Jun 2026 21:21 UTC

46 points

2 comments5 min readLW link

(blog.redwoodresearch.org)

CoT-forcing promptware

Bruce Middleton18 Jun 2026 19:33 UTC

2 points

0 comments2 min readLW link

AI that represents you can’t be neutral.

agulaya2418 Jun 2026 18:50 UTC

−5 points

2 comments3 min readLW link

On “Model Organisms”

J Bostock18 Jun 2026 18:42 UTC

18 points

1 comment6 min readLW link

Introduction: Gaussian Natural Latents

Haru18 Jun 2026 18:41 UTC

17 points

1 comment3 min readLW link

Contra Pace on When to Apologize

Zack_M_Davis18 Jun 2026 16:49 UTC

32 points

10 comments6 min readLW link

(zackmdavis.net)

Your Model Organisms Might Be Fried

Daniel Tan, J Bostock, draganover, ma-rmartinez, sidbaines and David Africa

18 Jun 2026 16:18 UTC

66 points

3 comments7 min readLW link

Shard narcissism as delusion of unembededness

Fernand018 Jun 2026 14:29 UTC

9 points

1 comment4 min readLW link

War of Dots: CRUSHING my opponents with FACTS and LOGIC

momom218 Jun 2026 12:07 UTC

17 points

2 comments7 min readLW link

How far do open weights trail the frontier?

RobinHa18 Jun 2026 11:01 UTC

23 points

3 comments1 min readLW link

(robinhaselhorst.com)

GLM 5.2 playing text adventures

kqr18 Jun 2026 7:23 UTC

13 points

1 comment1 min readLW link

(entropicthoughts.com)

Leveraged on being right

Ben Pace, the Vacationing Vagabond18 Jun 2026 6:51 UTC

40 points

6 comments3 min readLW link

Vulnerabilities and exploits: where are we headed?

tchauvin18 Jun 2026 5:49 UTC

9 points

0 comments5 min readLW link

(tchauvin.com)

Agents are under-elicited: A case study in optimization tasks

zef, kaivu, leni and rohuang

18 Jun 2026 2:39 UTC

17 points

1 comment7 min readLW link

(fulcrum.inc)

A preliminary experiment regarding consistency as a measure of conceptual abilities in language models

Chi Nguyen17 Jun 2026 22:56 UTC

16 points

3 comments7 min readLW link

(casparoesterheld.com)

Gears for political races

Tom Smith17 Jun 2026 20:19 UTC

133 points

14 comments14 min readLW link

“Did you lie?” Evaluating Lie Detectors across Model Scale and Belief-Verified Model Organisms

Alan Cooney, David Africa and Geoffrey Irving

17 Jun 2026 18:43 UTC

30 points

0 comments6 min readLW link

(arxiv.org)