Purely local autoregressive systems cannot maintain long-range coherence at finite temperature.

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

• Purely local-interaction systems on 1D topologies cannot maintain long-range order at finite temperature.claim0.832
• The inability for autoregressive large language models to maintain states of long-range order resembles tangential speech or derailment in formal thought disorder.claim0.783
  Analogy between LLM incoherence and schizophrenia symptoms
• Topological no-go theorem (Sacco-Sakthivadivel-Levin) will establish that autoregressive LLMs cannot maintain long-range coherence via local interactions alone within 6 months of widespread community prediction0.780
• Autoregressive language models cannot converge to single stored patterns beyond their context window from local interactions alone.claim0.780
• Autoregressive model unable to converge to a single stored pattern for any finite β (Corollary 2)finding0.767
  Consequence of Theorem 3 and 1D no-order result
• long-range coherenceconcept0.762
  Ability to maintain structural consistency over extended sequences
• Emotions are not strictly locally scoped but instead bursty with a long tail of slow change persisting over 100 tokensclaim0.752
  Characterizes the temporal dynamics of emotion feature activation in LLMs
• To what extent is there persistence of emotional state beyond what is expected merely from the autoregressive nature of LLMs?question0.751
  The central research question motivating the paper