Bayesian model outputs aggregate posterior 0.913 for a fly with indicator evidence concentrated at fine-grained (Level 4/5) levels

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

• Bayesian model outputs aggregate posterior 1.000 for a human with all 37 indicators activated, invariant to level-credence weightingfinding0.877
  Face-validity anchor case at the positive extreme.
• In simulations, positive evidence threshold for Bayesian model reduction corresponds to ΔF ≤ −3, equivalent to odds ratio of exp(−3) ≈ 0.05 (reduced model ~20 times more likely than full model).finding0.779
  Quantitative threshold used for accepting reduced models; linked to Bayes factor of ~20
• Models with as few as 14B parameters can exhibit 40% misalignment while maintaining 99% coherence (Turner et al. 2025 result)finding0.763
  Concurrent work result showing emergent misalignment occurs in small models
• Under reward shaping (G=100, H=-100, F=0), Active Inference scored 99.52, Bayesian RL 99.77, Q-learning 95.56, with nearly identical behavior between belief-based agents.finding0.761
  Table 2, row 3, showing equivalence when prior preferences match rewards.
• Bayesian model evidenceconcept0.758
  The probability of sensory data under a generative model; negative log evidence is bounded by free energy.
• Nearly all of 64 simulated agents attain 100% performance at around the 10th trial when allowed to perform abductive Bayesian model reduction after each trial.finding0.758
  Group-level simulation result showing generalizability of BMR benefit across agents
• In single-agent simulation of 32 trials, performance becomes perfect after trial 14 without Bayesian model reduction, with confidence increasing progressively.finding0.751
  Baseline learning curve for pure epistemic learning without structure learning
• Bayesian model reduction after 12 trials correctly removes off-diagonal (redundant) parameters from the likelihood array, recovering the true contingency structure.finding0.749
  Validation that BMR correctly identifies and prunes wrong connections in the likelihood mapping