Off-policy training data for LLM behavior probes causes significant generalization failures especially for intent-based behaviors like deception, and performance on coerced incentivised data correlates with real on-policy success.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.AI 3verdicts
UNVERDICTED 3representative citing papers
Persona axes derived from contrastive prompts and PCA yield linear probes that generalize better than raw-activation probes across 10 datasets for deception and sycophancy.
Dominant control-based AI alignment falls short for potential AGI subjects; a parenting model drawing on Turing's child machines should foster gradual autonomy and cooperative coexistence.
citing papers explorer
-
The Impact of Off-Policy Training Data on Probe Generalisation
Off-policy training data for LLM behavior probes causes significant generalization failures especially for intent-based behaviors like deception, and performance on coerced incentivised data correlates with real on-policy success.
-
Do Linear Probes Generalize Better in Persona Coordinates?
Persona axes derived from contrastive prompts and PCA yield linear probes that generalize better than raw-activation probes across 10 datasets for deception and sycophancy.
-
The Possibility of Artificial Intelligence Becoming a Subject and the Alignment Problem
Dominant control-based AI alignment falls short for potential AGI subjects; a parenting model drawing on Turing's child machines should foster gradual autonomy and cooperative coexistence.