Archive
Sequences
About
Search
Log In
Questions
Events
Shortform
Alignment Forum
AF Comments
Home
Featured
All
Tags
Recent
Comments
RSS
Refine
Tag
Last edit:
1 Sep 2022 14:27 UTC
by
NeuralBets
Refine
is a conceptual research incubator hosted by
Conjecture
.
Relevant
New
Old
confusion about alignment requirements
Tamsin Leake
6 Oct 2022 10:32 UTC
39
points
10
comments
3
min read
LW
link
(carado.moe)
Refine’s Second Blog Post Day
adamShimi
20 Aug 2022 13:01 UTC
19
points
0
comments
1
min read
LW
link
Benchmarking Proposals on Risk Scenarios
Paul Bricman
20 Aug 2022 10:01 UTC
25
points
2
comments
14
min read
LW
link
What if we approach AI safety like a technical engineering safety problem
zeshen
20 Aug 2022 10:29 UTC
33
points
4
comments
7
min read
LW
link
What I Learned Running Refine
adamShimi
24 Nov 2022 14:49 UTC
107
points
5
comments
4
min read
LW
link
PreDCA: vanessa kosoy’s alignment protocol
Tamsin Leake
20 Aug 2022 10:03 UTC
50
points
8
comments
7
min read
LW
link
(carado.moe)
the Insulated Goal-Program idea
Tamsin Leake
13 Aug 2022 9:57 UTC
43
points
4
comments
2
min read
LW
link
(carado.moe)
goal-program bricks
Tamsin Leake
13 Aug 2022 10:08 UTC
31
points
2
comments
2
min read
LW
link
(carado.moe)
ordering capability thresholds
Tamsin Leake
16 Sep 2022 16:36 UTC
27
points
0
comments
4
min read
LW
link
(carado.moe)
Refine Blogpost Day #3: The shortforms I did write
Alexander Gietelink Oldenziel
16 Sep 2022 21:03 UTC
23
points
0
comments
1
min read
LW
link
Refine’s Third Blog Post Day/Week
adamShimi
17 Sep 2022 17:03 UTC
18
points
0
comments
1
min read
LW
link
Representational Tethers: Tying AI Latents To Human Ones
Paul Bricman
16 Sep 2022 14:45 UTC
30
points
0
comments
16
min read
LW
link
Epistemic Artefacts of (conceptual) AI alignment research
Nora_Ammann
and
particlemania
19 Aug 2022 17:18 UTC
30
points
1
comment
5
min read
LW
link
Oversight Leagues: The Training Game as a Feature
Paul Bricman
9 Sep 2022 10:08 UTC
20
points
6
comments
10
min read
LW
link
Ideological Inference Engines: Making Deontology Differentiable*
Paul Bricman
12 Sep 2022 12:00 UTC
6
points
0
comments
14
min read
LW
link
Levels of goals and alignment
zeshen
16 Sep 2022 16:44 UTC
27
points
4
comments
6
min read
LW
link
Cataloguing Priors in Theory and Practice
Paul Bricman
13 Oct 2022 12:36 UTC
13
points
8
comments
7
min read
LW
link
Refine: what helped me write more?
Alexander Gietelink Oldenziel
25 Oct 2022 14:44 UTC
12
points
0
comments
2
min read
LW
link
Embedding safety in ML development
zeshen
31 Oct 2022 12:27 UTC
24
points
1
comment
18
min read
LW
link
A newcomer’s guide to the technical AI safety field
zeshen
4 Nov 2022 14:29 UTC
42
points
3
comments
10
min read
LW
link
Interlude: But Who Optimizes The Optimizer?
Paul Bricman
23 Sep 2022 15:30 UTC
15
points
0
comments
10
min read
LW
link
Summary of ML Safety Course
zeshen
27 Sep 2022 13:05 UTC
7
points
0
comments
6
min read
LW
link
My Thoughts on the ML Safety Course
zeshen
27 Sep 2022 13:15 UTC
50
points
3
comments
17
min read
LW
link
(Structural) Stability of Coupled Optimizers
Paul Bricman
30 Sep 2022 11:28 UTC
25
points
0
comments
10
min read
LW
link
Refine’s First Blog Post Day
adamShimi
13 Aug 2022 10:23 UTC
55
points
3
comments
1
min read
LW
link
Boolean Primitives for Coupled Optimizers
Paul Bricman
7 Oct 2022 18:02 UTC
9
points
0
comments
8
min read
LW
link
my current outlook on AI risk mitigation
Tamsin Leake
3 Oct 2022 20:06 UTC
63
points
6
comments
11
min read
LW
link
(carado.moe)
How I think about alignment
Linda Linsefors
13 Aug 2022 10:01 UTC
31
points
11
comments
5
min read
LW
link
Steelmining via Analogy
Paul Bricman
13 Aug 2022 9:59 UTC
24
points
0
comments
2
min read
LW
link
(paulbricman.com)
I missed the crux of the alignment problem the whole time
zeshen
13 Aug 2022 10:11 UTC
53
points
7
comments
3
min read
LW
link
All the posts I will never write
Alexander Gietelink Oldenziel
14 Aug 2022 18:29 UTC
53
points
8
comments
8
min read
LW
link
Refine: An Incubator for Conceptual Alignment Research Bets
adamShimi
15 Apr 2022 8:57 UTC
144
points
13
comments
4
min read
LW
link
How to Diversify Conceptual Alignment: the Model Behind Refine
adamShimi
20 Jul 2022 10:44 UTC
87
points
11
comments
8
min read
LW
link
No comments.
Back to top