Within-agent score standard deviation suggests deck order matters more than seat position.

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

• Within-agent score std exceeds cross-seat win-rate differentials by 1–2 orders of magnitudefinding0.820
  deck-order variance dominates seat-position variance
• Smith & Boyd (1991) criteria are irrelevant across the vast majority of the space of possible agents.claim0.760
  Dismissal of earlier criteria as too narrow.
• The code-agent ordering (TrackerAgent > SetRaceAgent > EconomyAgent) shows information exploitation matters more than greedy quartet-chasing, which in turn outperforms conservative budgetingclaim0.748
  interpretation of what drives success among deterministic strategies
• G3-F within-agent score std 4,026 on median 5,250finding0.735
  High score variance driven by deck order.
• Agentic self-evaluation and self-steering may scale to broadly interpret and understand internal representations and SAE features.claim0.735
  Forward-looking claim about the potential of model introspection as an interpretability tool
• Active inference agent with learnable preferences developed a strict preference for goals (score +) when the Frisbee location was encountered first, becoming a goal-seeking agent.finding0.734
  Figure 5.4 and text.
• The order relation ≤ on a poset P is determined entirely by the down-sets in P.claim0.731
• Whether the advantage of agentic self-evaluation over textual evaluation in predicting persistence is due to introspection quality or to testing of additional (including negative) steering strengthsquestion0.730
  Identified methodological gap in interpreting the self-evaluation experiment results