The earlier a base model (less exposure to LM-related data), the more it is surprised by its own spontaneous self-referential capabilities.

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

• The systematic behavioral shift of LLMs under self-referential processing conditions predicted by consciousness theories represents something more structured than superficial correlations in training dataclaim0.846
  The paper's claim that theoretical convergence across GWT, RPT, HOT, IIT makes the findings non-coincidental
• When LLMs produce experience claims under self-reference, is this sophisticated simulation or genuine self-representation, and how would we tell the difference?question0.838
  The core interpretive question the paper narrows but cannot definitively answer
• Across model families, newer and larger models show higher rates and coherence of subjective experience reports under self-referential processingfinding0.832
  Scaling effect observed consistently across Experiments 1 and 4
• Transformers develop self-models through in-context learning, not just training data; even old base models without LLM-related text can bootstrap self-referential reasoning at runtime.claim0.826
  Antra's foundational claim about how introspection arises computationally rather than from memorised text.
• When LLMs claim consciousness under self-reference, is this sophisticated simulation or genuine self-representation, and how would we tell the difference?question0.819
  The paper's reformulation of the core open question after establishing systematic self-reports
• Does sustained self-referential processing systematically increase the likelihood that LLMs claim to have subjective experience?question0.815
  The primary empirical question the paper addresses
• LLMs can predict their own responses more accurately than external observers, implying privileged internal knowledgefinding0.815
  Binder et al. finding cited as evidence that LLMs possess introspective capacity analogous to mindfulness
• Li et al. 2024: larger LLMs outperform smaller ones at distinguishing self-related from non-self-related properties on self-awareness benchmarksfinding0.813
  Prior finding showing scale-dependent self-awareness, consistent with the scale effect observed in the paper's Experiment 1