Philosophical vocabulary is negatively correlated with scores in contemplative condition (model-level r=-0.72)

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

• A 337-character contemplative system prompt lifts all 28 models by +2.62 points on a 10-point scale.finding0.765
  Core empirical result: every model, every architecture, every alignment type responds to the contemplative prompt with measurable gain.
• Contemplative practice progressively opacifies this constraint by developing a model of the agent's own QRF dynamics, revealing the partition as a contingent modelling choice rather than a given feature of reality.claim0.760
  Mechanism of contemplative training.
• The inability for autoregressive large language models to maintain states of long-range order resembles tangential speech or derailment in formal thought disorder.claim0.754
  Analogy between LLM incoherence and schizophrenia symptoms
• Response length (words) correlates with scores at r=0.22 baseline and r=0.12 contemplative; explains only ~5% of variancefinding0.753
  Discriminant validity: composite scores are not reducible to verbosity
• Contemplative prompt elevates self-observation task performance in language models.finding0.749
  Supports Janus's claim that introspection is architecturally available; prompting determines whether/how capacity is leveraged.
• Mean validated introspective fidelity across concept-model pairs: R²=0.12 (1B), 0.37 (3B), 0.61 (8B); pooled LMM β=0.29, p=5.55×10⁻⁹⁹finding0.749
  Strong scaling trend for introspective fidelity when excluding invalid steering-sign pairs
• Any deepening of an LLM's linguistic understanding of contemplative principles as it scales may enhance the effectiveness of CCAI and CRL approacheshypothesis0.747
  Scaling hypothesis for language-based contemplative alignment approaches
• Our results demonstrate that modern language models possess at least a limited, functional form of introspective awareness.quote0.745
  Abstract's main conclusion.