The model tends to reflect more when the question is difficult, and accuracy is generally lower for harder questions

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

• Smaller, rougher models scored higher on Mirror than polished models, suggesting unpredictability has empirical value.claim0.823
• Larger models should amplify bias less than smaller models, with model biases more accurately reflecting data biases rather than exacerbating themclaim0.823
  Implication of PRH for AI fairness and bias
• Roughness in responses decreases with parameter count within same-alignment model families, operationalizing the cost of polishing.claim0.804
• Bigger models are more likely to converge to a shared representation than smaller modelshypothesis0.797
  Selective pressure toward convergence via model capacity
• How does reflection influence the model's reasoning performance?question0.793
  Second central research question motivating ReflCtrl investigation
• Within each difficulty category, correctness rate is not correlated with reflection rate, suggesting reflection may be redundantclaim0.790
  Per-category analysis showing reflection rate does not help within difficulty class
• Earlier/less capable models exhibit a larger gap between think and don't think representation strengthfinding0.786
  Claude 3 models show a bigger difference than newer models like Opus 4.1.
• We hypothesize that explicitly instructing the model to evaluate the correctness of the given statement may change the geometry of truth directions.hypothesis0.785
  Motivating hypothesis for Section 5's investigation of prompt template effects.