The budget term −log k acts as a regularizer to discourage degenerate long prompts.

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

• Peter Landin recommended 'denotative' to replace ill-defined 'functional' and 'declarative'.claim0.751
  Historical basis for denotational design.
• UCCT fills a gap in explaining when behavior flips for a specific prompt and how much anchor budget is neededclaim0.742
  Authors contrast their work with prior phase/representation studies
• Fine-tuning reduces dr; retrieval increases effective ρd; few-shot k trades budget against bothhypothesis0.732
  UCCT's unified view of adaptation methods
• Optimal number of features scales faster than optimal number of training steps with compute budget.finding0.729
  Allocation result from scaling laws.
• The formal concepts with frequent intent form an order filter in the concept lattice, called the iceberg.claim0.729
  Definition/claim about iceberg lattices.
• A question anywhere along the line that elicits a premature attempt at an answer could neutralize the remainder of the process into rationalization.hypothesis0.727
  About chain-of-thought and process safety.
• Length normalization prevents degenerate tool-calling trajectories and repeated tool calls without normalization.finding0.727
  Empirical result showing that without length normalization, RL training produces rapidly increasing tool usage with performance collapse and repetitive tool calls.
• Random word prefix prompts show emergence patterns similar to no-prompt, suggesting prompt length alone does not shift truth geometry.claim0.725
  Control experiment ruling out token-count as the cause of truth geometry shifts.