Self-correcting search improves viable candidate success rate from 6.5% to ~30% (4.6x improvement)

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

• Self-correcting search is Pareto-optimal across tested conditioning strengths.claim0.815
  Asserts that the method maintains efficiency across a range of constraint strengths without degradation.
• Self-Correcting Searchmethod0.765
  Technique using internal model representations as feedback loops to steer diffusion-based materials generation toward target properties.
• Fine-tuning Llama-3.1-8B on self-correction examples increases multi-attempt rate proportionally with training data ratiofinding0.756
  Shows behavioral pattern of self-correction is trainable in smaller models
• Pearson-Vogel et al.: accurate self-description prompts increase introspective detection from 0.3% to 39.9%finding0.750
  Cited to mechanistically support why the contemplative prompt changes what post-training-shaped final layers allow through
• Self-correcting search with interpretability feedbackconcept0.745
• Self-correcting search employs the same conceptual move as Wurgaft's manifold steering, applied to chemistry instead of LMsclaim0.744
  Interpretive assertion that the internal-state feedback mechanism mirrors manifold steering from prior work.
• Within each difficulty category, correctness rate is not correlated with reflection rate, suggesting reflection may be redundantclaim0.735
  Per-category analysis showing reflection rate does not help within difficulty class
• Self-referential processing yields significantly higher self-awareness scores than conceptual control on paradoxical reasoning: t(399)=14.90, p=3.0×10⁻⁴⁰finding0.732
  Experiment 4 result ruling out semantic priming as explanation for the experimental effect