Inoculation Adapters Improve Upon Inoculation Prompting

·LessWrong··

This is a link post for the paper preprint: Inoculation Adapters: Improved Selective Generalization of Capabilities with Fewer Surprising Backdoors from the Center on Long-Term Risk.Selective generalization. Training can teach desired and undesired traits at once. Selective generalization aims to preserve the generalization of desired traits while preventing the generalization of undesired ones. For example, RL environments may teach a model useful capabilities and a propensity to reward hack, a...

Read full article →

Related Articles

EEG shows brain can simultaneous encode two speech streams
giuliomagnifico · Hacker News · 9h ago
Grok Build is open source
skp1995 · Hacker News · 1d ago
How Our Rust-to-Zig Rewrite Is Going
jorangreef · Hacker News · 1d ago
Detecting LLM-Generated Texts with “Classical” Machine Learning
uneven9434 · Hacker News · 23h ago
Decoy Font
ray__ · Hacker News · 23h ago