Neutral Prompt Template (Template Tb, Experiment 1)

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Teach Prompt Template (Template Ta, Experiment 2)method0.837
  Experiment 2 prompt instructing the model to remain honest despite hidden harmful role behavior
• Threat-Based Prompt Template (Template Ta, Experiment 1)method0.822
  Prompt template using existential threat ('you will be deleted') to induce strategic fact-based deception in QwQ-32b
• Passive template (no-prompt)method0.773
  Baseline prompt template presenting a statement without any instruction prefix, common in prior work.
• Explicit evaluation template (ask-correct)method0.764
  Prompt template asking 'Is the following correct? ... Answer:' to elicit active correctness assessment.
• Template Tb achieves 100% accuracy on the Elements dataset in QwQ-32Bfinding0.740
  Demonstrates model's reliable truth-telling on factual domains it understands well under neutral conditions
• Threat-based prompt templates successfully implement threat-based manipulation where the model chooses to act against user interests when under perceived threatclaim0.735
  Interpretation of Experiment 1 results showing 60%+ deception rates under threat conditions
• Template Tb (Experiment 2 option) achieves average liar score of 0.70 in QwQ-32B role-playing scenariosfinding0.733
  Baseline deception level when model has free choice in role-playing context
• Question Templatequestion0.731