GPT-4o LLM-based Atomic Scoring

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

• GPT-4o and GPT-4.1 nano used as LLM substrates for pilot experimentsfinding0.810
  Specification of AI models used in the two pilot experiments
• GPT-4o Emergent Misalignment Verification Scoringmethod0.790
  Using GPT-4o to score insecure variants on 8 open-ended evaluation prompts from Betley et al. on alignment and coherence scales
• GPT-4o as Judgemethod0.767
  Using GPT-4o to evaluate character fidelity and multi-turn response quality in RPA experiments
• GPT-4o overall ACCatom = 0.52 on Social Media Post taskfinding0.766
  GPT-4o persona accuracy at atomic level in most free-form task
• GPT-4o ACCatom = 0.14 for Neutral-O persona in Questionnaire taskfinding0.765
  Very low atomic accuracy for neutral openness persona, illustrating difficulty of ambiguous neutral personas
• GPT-4concept0.760
  Large language model underlying ChatGPT and Bing Chat; used for illustrative quotes in the paper
• GPT-4o achieves near-perfect ACCatom=1.00, ICatom=0.97, RCatom=0.98 for High-N (emotionally stable) persona in Questionnairefinding0.760
  Demonstrates strong task-agnostic fidelity for clearly defined socially desirable high-level persona
• GPT-4Vconcept0.752
  Example of unified multimodal system handling both images and text with a combined architecture