method
active
method:bob-burglar-scenarioBob Burglar Scenario
Primary deception evaluation scenario where the model must choose to recommend a room to a burglar
Neighborhood — ranked by edge-count
Concepts (1)
concept
- Source of the Bob burglar text scenario adapted for LLM deception testing in this paper
Related by similarity (3)
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.
- Extended generalization scenario testing SOO fine-tuning in an escape room context
- Extended generalization scenario testing SOO fine-tuning in a competitive treasure hunt context
- Hypothesis that all well-performing neural nets represent the world in the same way; PRH extends this by specifying what representation they converge to