hypothesis
active
hypothesis:both-models-must-first-learn-a-behavioral-prior-before-privileged-signal-can-be-learnedBoth models must first learn a behavioral prior before privileged signal can be learned
Explains absent privileged access.
Source paper
extracted_from(2026) · Adam Karvonen · Euan Ong · Subhash Kantamneni · Samuel Marks
Neighborhood — ranked by edge-count
Papers (1)
paper
Findings (1)
finding
- No evidence of privileged access.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- The paper's central mechanistic explanation of why narrow fine-tuning causes broad misalignment
- During pretraining the model learns distinct harm categories and during finetuning wires them into general harmful-request features.hypothesis0.772Suggested by dictionaries trained on pretraining data having few refusal features.
- Developmental analogy used to explain sample efficiency under high ρd conditions
- Claim that capability emerges from architecture, not data, and that later models lose the surprise.
- Can active inference agents learn their own prior preferences without explicit reward signals?question0.761Question answered by the preference learning experiments.
- Core goal statement.
- Predicted neural signature of insight: reduced ERP latency and increased early amplitude
- Hypothesis box in §3.2.