Does autoregressive use of transformer-based LLMs constitute algorithmic recurrence for RPT-1, or does it depend on where we draw the system boundary?

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

• The problem isn't that it is a transformer. The problem is that it is an auto-regressive LLM. Auto-regressive LLMs that compute each token with a fixed number of computational steps can't reason, regardless of the details of the architecture.quote0.800
  LeCun's post on X supporting the view that fixed-step probabilistic prediction precludes consciousness in LLMs.
• RPT-1: Input modules using algorithmic recurrenceconcept0.800
  Indicator derived from RPT: use of algorithmic recurrence in input modules.
• autoregressive recurrenceconcept0.791
  Transformers are recurrent through autoregression because the K/V stream provides horizontal information flow across positions, even though each forward pass is feedforward.
• Transformers are recurrent through autoregression because K/V stream provides horizontal information flow across positions.claim0.788
  Claim formalizing the Anima Labs idea that transformers are effectively recurrent due to K/V stream.
• RPT-1 (algorithmic recurrence) may be useful as a sensitive indicator whose absence is strong evidence that a system is not conscious.claim0.783
  Illustrates how even simple AI properties can be informative as sensitive negative-direction indicators
• Autoregressive language models cannot converge to single stored patterns beyond their context window from local interactions alone.claim0.767
• Algorithmic recurrence is likely necessary for conscious experience with human-like temporal character.claim0.762
  Support for RPT-1.
• To what extent is there persistence of emotional state beyond what is expected merely from the autoregressive nature of LLMs?question0.749
  The central research question motivating the paper