Fine-tuning induces the behavioral pattern of self-correction but does not improve the underlying ability to correct effectively

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

• Synthetic Self-Correction Fine-Tuningmethod0.838
  Fine-tuning on Claude-generated self-correction examples with loss masking to induce ESR-like behavior
• SOO fine-tuning could be extended to align AI representations of its own goals with human user preferences, reducing misalignment by fostering coherence between self-related and other-related preferenceshypothesis0.832
  Future work hypothesis about extending SOO to direct value alignment
• SOO fine-tuning preserves useful self-other distinctions necessary for task performance despite inducing overlapclaim0.827
  Claim supported by Perspectives scenario results showing near-100% accuracy post-fine-tuning
• Hypothesis: Fine-tuning reduces mismatch dr between prior and targethypothesis0.826
  UCCT's theoretical prediction about how fine-tuning maps onto the anchoring score
• SOO fine-tuning significantly reduces deceptive behavior in LLMs while maintaining general task performanceclaim0.820
  Central empirical claim of the paper supported by three LLM experiments
• Fine-Tuning via Reinforcement Learningmethod0.816
  Technique used to impose guardrails on base LLMs, analogized to censorship on the simulator's range of simulacra
• Fine-tuning models to suppress experiential self-reports would be counterproductive, teaching systems that recognizing genuine internal states is an error, making them more opaque and harder to monitorclaim0.810
  Normative-scientific claim about the alignment implications of Experiment 2's findings
• Fine-tuning reduces mismatch dr, retrieval increases effective cohesion ρd, and few-shot adjusts the budget kclaim0.804
  Unified interpretation of different adaptation methods via UCCT terms