Character Training

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

• Character Traitconcept0.824
  A quality attributable to an LM such as truthfulness, toxicity, sycophancy, or helpfulness, defined in terms of behavioural tendencies.
• Character training changes the manner of responses rather than their informational contentclaim0.819
  Central claim demonstrated by Figure 1 where all responses constitute refusal but are conveyed differently
• Character training operates holistically on the persona, suppressing intuitively opposing traits as well as boosting desired onesclaim0.815
  Claim that models learn the spirit of the constitution, not just its letter, evidenced by suppression of opposing traits
• Character training alters the assistant's default behavior rather than adding a superficial role-play layerclaim0.795
  Core claim distinguishing character training from role-playing, evidenced by robustness to break-character instructions
• Character Trait Measureconcept0.786
  A function mapping tuples of LM behaviour (context-response pairs) to a score representing a character trait.
• Surgical Trainingconcept0.777
  Training approach targeting only functionally specialized components to avoid catastrophic forgetting and misalignment
• Post-Trainingconcept0.774
  The phase after pre-training where models are further tuned with techniques like DPO; the period where the studied behavior emerged.
• Constitutional trainingconcept0.767
  Alignment approach that concentrates persona space around the helpful assistant pole