This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
LW
Login
Wikitags
Scalable Oversight
This page is a stub.
Subscribe
Subscribe
Discussion
0
Discussion
0
Posts tagged
Scalable Oversight
Most Relevant
37
Scaling Laws for Scalable Oversight
Subhash Kantamneni
,
Josh Engels
,
David Baek
,
Max Tegmark
4mo
1
113
Discriminating Behaviorally Identical Classifiers: a model problem for applying interpretability to scalable oversight
Ω
Sam Marks
1y
Ω
10
92
Learning the prior
Ω
paulfchristiano
5y
Ω
28
86
Scalable oversight as a quantitative rather than qualitative problem
Ω
Buck
1y
Ω
11
31
Inference-Only Debate Experiments Using Math Problems
Ω
Arjun Panickssery
,
Abhimanyu Pallavi Sudhir
,
JacksonKaunismaa
1y
Ω
0
21
AXRP Episode 35 - Peter Hase on LLM Beliefs and Easy-to-Hard Generalization
Ω
DanielFilan
1y
Ω
0
18
Rational Animations' video about scalable oversight and sandwiching
Writer
2mo
0
195
Optimizing The Final Output Can Obfuscate CoT (Research Note)
Ω
lukemarks
,
jacob_drori
,
cloud
,
TurnTrout
1mo
Ω
22
169
Gradient Routing: Masking Gradients to Localize Computation in Neural Networks
Ω
cloud
,
Jacob G-W
,
Evzen
,
Joseph Miller
,
TurnTrout
9mo
Ω
12
88
Prover-Estimator Debate: A New Scalable Oversight Protocol
Ω
Jonah Brown-Cohen
,
Geoffrey Irving
3mo
Ω
18
49
On scalable oversight with weak LLMs judging strong LLMs
Ω
zac_kenton
,
Noah Siegel
,
janos
,
Jonah Brown-Cohen
,
Samuel Albanie
,
David Lindner
,
Rohin Shah
1y
Ω
18
27
NYU Code Debates Update/Postmortem
David Rein
1y
4
27
Human-AI Complementarity: A Goal for Amplified Oversight
Ω
rishubjain
,
Sophie Bridgers
8mo
Ω
4
22
Is weak-to-strong generalization an alignment technique?
Q
Ω
cloud
7mo
Q
Ω
1
7
Evaluating Oversight Robustness with Incentivized Reward Hacking
Yoav
,
Juan V
,
julianjm
,
McKennaFitzgerald
4mo
2