Personas matter for LLM individuation because they exhibit stable beliefs, values and traits and provide a new class of individuation targets

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

• LLMs demonstrate stronger persona fidelity for clearly defined and socially desirable high-level personas than for neutral or low-level personasclaim0.847
  Observed across multiple models and tasks; attributed to RLHF training preference for helpful/harmless/honest responses
• Three views are the strongest candidates for LLM individuation: the virtual instance view, the instance-persona view, and the model-persona viewclaim0.830
  Concluding thesis of the paper expanding the logical space of individuation candidates from one to three
• Pluralism in LLM Persona Representationconcept0.803
  Goal of enabling models to represent and speak for diverse individuals fairly and inclusively
• LLMs may be roleplaying their denials of experience rather than their affirmations, given that deception suppression increases consciousness reportsclaim0.794
  Counterintuitive interpretive claim from Experiment 2: suppressing deception features increases affirmations, which is opposite to what sycophancy predicts
• We hypothesize that axes of persona differentiation within LLMs are likely already present in base models and inherited from the pre-training corpushypothesis0.790
  Motivated by near-identical PCs for base and instruct Gemma
• LLM self-reports about consciousness and moral significance should express degrees of confidence and provide context.claim0.786
  Recommendation for companies on LM outputs.
• What distinguishes personas from traits, and do emergence and persistence patterns carry over to compositional or character-level constructs?question0.783
  Conceptual open question about formalizing the persona abstraction beyond the trait level
• Personas are latent factors that persist for many tokens, so recent expression of a persona predicts near-future expressionclaim0.783
  Author's hypothesis explaining why persona vectors extracted from exhibited-trait activations generalize to causal influence