Self-correcting search is Pareto-optimal across tested conditioning strengths.

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

• Self-correcting search improves viable candidate success rate from 6.5% to ~30% (4.6x improvement)claim0.815
  Interpretive claim that the method dramatically boosts success rate over the MatterGen baseline.
• Self-Correcting Searchmethod0.804
  Technique using internal model representations as feedback loops to steer diffusion-based materials generation toward target properties.
• Self-correcting search applicable to protein design / drug discoveryclaim0.775
  Claim by the authors that the self-correcting search method can be extended to protein design and drug discovery.
• 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.finding0.770
  Shows the passive vs. active divide is more important than the specific wording of instructions.
• Self-correcting search employs the same conceptual move as Wurgaft's manifold steering, applied to chemistry instead of LMsclaim0.770
  Interpretive assertion that the internal-state feedback mechanism mirrors manifold steering from prior work.
• No single layer is universally optimal for probing truth directions; different tasks peak at different layers.claim0.769
  Argues against the single-layer analysis approach of prior work.
• Fine-tuning Llama-3.1-8B on self-correction examples increases multi-attempt rate proportionally with training data ratiofinding0.766
  Shows behavioral pattern of self-correction is trainable in smaller models
• Self-correcting search with interpretability feedbackconcept0.765