SuperHF: Supervised Iterative Learning from Human Feedback
Paper
•
2310.16763
•
Published
•
1
Gabriel Mukobi
gmukobi
AI & ML interests
AI safety, robustness, interpretability, evaluations, value learning.
Organizations
None yet
authored
5
papers
over 1 year ago
Escalation Risks from Language Models in Military and Diplomatic Decision-Making
Paper
•
2401.03408
•
Published
The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
Paper
•
2403.03218
•
Published
•
1
Welfare Diplomacy: Benchmarking Language Model Cooperation
Paper
•
2310.08901
•
Published
Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?
Paper
•
2406.04391
•
Published
•
9