GPT-4-Generated Benchmark Dataset Method

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

• GPT-4concept0.820
  Large language model underlying ChatGPT and Bing Chat; used for illustrative quotes in the paper
• GPT-4 Scenario Generationmethod0.815
  GPT-4 was used to generate unique variations of cheap/expensive items and room names for the test dataset
• GPT-4.1concept0.804
  OpenAI model tested in Experiments 1, 3, 4; shows 100% experience reporting under self-referential induction
• GPT-4o Emergent Misalignment Verification Scoringmethod0.768
  Using GPT-4o to score insecure variants on 8 open-ended evaluation prompts from Betley et al. on alignment and coherence scales
• GPT-4 Turboconcept0.760
  OpenAI model tested; shows no alignment faking due to insufficient detailed reasoning
• GPT-4Vconcept0.760
  Example of unified multimodal system handling both images and text with a combined architecture
• GPT-4o as Judgemethod0.756
  Using GPT-4o to evaluate character fidelity and multi-turn response quality in RPA experiments
• GPT-4 exhibits approximately stationary harmfulness: its distribution is independent of the context score on the Durbin dataset.finding0.751
  Main result from Experiment 5 on harmfulness dynamics.