tag
reward-hacking
; 7 entries
August 2026
◐
23
Anthropic Risk Report: August 2026
◐
22
How do cars do in out-of-sample crash testing?
12
You're Absolutely Right
11
Four LLM loss functions → four flavors of LLM misalignment
5
Why don't we just give AI the answers?
5
The Most Forbidden Technique
3
We need to RL less