Training natural schemers, model organisms of collusion, scaling laws for safety probes, ...

Also: red-teaming multi-agent attacks for loss-of-control risks, why static probes fail for deception, towards robust model organisms for eliciting latent knowledge, predicting the onset of detectable reward hacking before RL training, neural networks are already clean, subliminal learning can transfer learned capabilities and backdoors

saved by

• Bhagyesh Kumar