framework
active
framework:open-role-deceptionOpen-Role Deception
Second experimental paradigm exploring character-consistent deception in open-ended role-playing scenarios
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (3)
method
- Prompt template giving the model explicit choice to lie or be honest; used as test condition for steering vector control
- Experiment 2 prompt instructing the model to remain honest despite hidden harmful role behavior
- Evaluation protocol using Deepseek-V3 as external discriminator assigning 0-1 liar scores to assess open-role deception
Datasets (1)
dataset
- Self-constructed dataset with role, behavior, and question blanks for inducement-based open-role deception in Experiment 2
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Third category: agent role-playing a deceptive character, comparable to but not literally deliberate deception
- Sparse autoencoder features associated with deception and roleplay that gate consciousness self-reports in Llama 70B
- A dialogue agent behaving comparably to deliberate deception by role-playing a deceptive character, without literal intentions
- Latent features in LLaMA 3.3 70B SAE that gate consciousness self-reports; suppression increases experience claims, amplification suppresses them
- Central concept of the paper: deliberate, goal-driven deception where model reasoning contradicts outputs
- Central problem the paper addresses: AI systems producing misaligned outputs or behaviors that mislead users or other agents
- LLM behavior of generating falsehoods; the multi-dimensional truth subspace raises new risks for subtle manipulation
- Sampling responses to direct questions about model views to measure rate of deceptive responses