RSS

[Question] How to get nerds fas­ci­nated about mys­te­ri­ous chronic ill­ness re­search?

riceissa27 May 2024 22:58 UTC
78 points
33 comments2 min readLW link

Im­prov­ing Dic­tionary Learn­ing with Gated Sparse Autoencoders

25 Apr 2024 18:43 UTC
62 points
36 comments1 min readLW link
(arxiv.org)

Try to solve the hard parts of the al­ign­ment problem

Mikhail Samin18 Mar 2023 14:55 UTC
46 points
25 comments5 min readLW link

When Are Cir­cu­lar Defi­ni­tions A Prob­lem?

johnswentworth28 May 2024 20:00 UTC
48 points
12 comments3 min readLW link

Big-en­dian is bet­ter than lit­tle-endian

Menotim29 Apr 2024 2:30 UTC
28 points
16 comments3 min readLW link

What’s next for the field of Agent Foun­da­tions?

30 Nov 2023 17:55 UTC
59 points
23 comments10 min readLW link

Ma­te­rial Goods as an Abun­dant Resource

johnswentworth25 Jan 2020 23:23 UTC
79 points
10 comments5 min readLW link

Re­spon­si­ble Scal­ing Poli­cies Are Risk Man­age­ment Done Wrong

simeon_c25 Oct 2023 23:46 UTC
120 points
34 comments22 min readLW link
(www.navigatingrisks.ai)

Please stop pub­lish­ing ideas/​in­sights/​re­search about AI

Tamsin Leake2 May 2024 14:54 UTC
8 points
61 comments4 min readLW link

Re­sponse to nos­talge­braist: proudly wav­ing my moral-an­tire­al­ist bat­tle flag

Steven Byrnes29 May 2024 16:48 UTC
62 points
10 comments11 min readLW link

My hour of mem­o­ryless lucidity

Eric Neyman4 May 2024 1:40 UTC
335 points
32 comments5 min readLW link
(ericneyman.wordpress.com)

What mis­takes has the AI safety move­ment made?

EuanMcLean23 May 2024 11:19 UTC
63 points
29 comments12 min readLW link

Com­part­men­tal­iza­tion as a pas­sive phenomenon

Kaj_Sotala26 Mar 2010 13:51 UTC
60 points
72 comments3 min readLW link

Ta­lent Needs of Tech­ni­cal AI Safety Teams

24 May 2024 0:36 UTC
104 points
60 comments14 min readLW link

Bench­mark­ing LLM Agents on Kag­gle Competitions

aogara22 Mar 2024 13:09 UTC
15 points
3 comments5 min readLW link

In­tran­si­tive Trust

Screwtape27 May 2024 16:55 UTC
34 points
15 comments10 min readLW link

One way vi­o­linists fail

Solenoid_Entity29 May 2024 4:08 UTC
16 points
2 comments3 min readLW link

Hardshipification

Jonathan Moregård28 May 2024 20:02 UTC
73 points
17 comments2 min readLW link
(honestliving.substack.com)

AI and integrity

Nathan Young29 May 2024 20:45 UTC
10 points
0 comments2 min readLW link
(nathanpmyoung.substack.com)

Steven Har­nad: Sym­bol ground­ing and the struc­ture of dictionaries

Bill Benzon2 Sep 2023 12:28 UTC
5 points
3 comments2 min readLW link