Enacted reflection may correspond to silent mid-layer processing; described reflection to the motor impulse of concepts leaking through to output.

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

• Reflection is not merely a behavioral artifact of prompting but a phenomenon encoded in the model's activation space.claim0.810
  Central interpretive claim of the paper, supported by steering vector experiments.
• Enacted Reflectionconcept0.800
  Responses that perform the observing act; contrasted with described reflection; scorer rewards enacted over described
• Reflective reasoning requires late-stage integration of semantic and reasoning signals, hence reflection-related directions emerge more clearly in higher network layers.claim0.778
  Interpretive claim about the locus of reflection in transformer architecture.
• Contrasting No Reflection with Triggered Reflection (µ(0→2)) provides a stronger reflection signal than contrasting Intrinsic with Triggered Reflection (µ(1→2)).claim0.778
  Empirical interpretation of which reference baseline yields more useful steering vectors.
• Suppressing reflection is considerably easier than inducing it, because inhibition requires the model to terminate reasoning while enhancement demands additional cognitive effort to re-examine reasoning trajectories.claim0.753
  Key asymmetry finding interpreted mechanistically by the authors.
• A linear reflection direction exists in reasoning LLMs' latent representation space that governs self-reflection behaviorclaim0.753
  Core claim of ReflCtrl that a single direction captures and controls reflection
• Reflection does not only emerge in SFT or RL stages but arises earlier during pre-training.claim0.751
  Cited finding from Shah et al. contextualizing the training origins of reflection.
• The inner voice is stilled, muffled, and there is hardly any possibility to cry out against ugliness.claim0.751