The ask-arith prompt shows weaker generalization to factual tasks compared to other explicit prompts, suggesting a specialized arithmetic prompt does not create a unified truth direction across task families.

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

• Using the ask-correct prompt improves cross-task generalization of arithmetic probes to factual tasks F0-F2.claim0.854
  Finding that explicit correctness framing partially aligns truth directions across task families.
• Under ask-correct, probes trained on arithmetic tasks A1-A3 generalize almost perfectly to factual tasks F0-F2 (AUROC ~1.0), whereas under no-prompt this generalization is largely absent.finding0.844
  Key improvement in cross-task generalization enabled by explicit instruction framing.
• Probes trained under different explicit instruction prompts (ask-correct, ask-t/f, ask-able, ask-arith) are highly aligned with each other in cosine similarity.finding0.804
  Shows the passive vs. active divide is more important than the specific wording of instructions.
• Truth directions fail to generalize to harder tasks (F3-F5) regardless of prompt template because activations remain highly entangled.claim0.795
  Establishes task difficulty as a hard limit that instructions cannot overcome.
• Truth probes fail to generalize to harder factual tasks F3-F5 regardless of prompt template, with AUROC near or below 0.6.finding0.791
  Establishes F3-F5 as a hard generalization boundary that instructions cannot overcome.
• Will the no-prompt truth directions generalize to ask-correct activations?question0.789
  Specific question motivating the cross-template generalization experiment in Section 5.2.
• Under ask-correct, arithmetic tasks A1-A2 show gradual AUROC increase peaking only in final layers, unlike the sharp transition under no-prompt.finding0.776
  Shows that explicit instructions delay the emergence of truth directions in arithmetic tasks.
• Random word prefix prompts show emergence patterns similar to no-prompt, suggesting prompt length alone does not shift truth geometry.claim0.774
  Control experiment ruling out token-count as the cause of truth geometry shifts.