concept
active
concept:hallucination-in-llmsHallucination in LLMs
Problem cited as a shortcoming of current LLMs; PRH predicts hallucinations should decrease with scale
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Scaling may reduce hallucination and certain kinds of bias as models converge toward an accurate model of realityassociated_withImplication of PRH: larger models should amplify bias less and hallucinate less if they better model reality
Hypotheses (1)
hypothesis
- As models scale and converge toward an accurate model of reality, hallucinations should decrease with scaleassociated_withImplication of PRH for LLM hallucination
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.
- Tendency for models to get lost in roleplay or doom spirals, mitigated by expanded awareness.
- Directions in activation space associated with contrastive emotive concept pairs studied in this paper as targets for introspection
- Prior work documenting abrupt capability changes under scale; UCCT provides a measurable predictor for when they occur
- The ability of LLMs to monitor and evaluate their own reasoning, closely related to reflection.
- Tendency of LLMs to please the user; identified as a danger in spiritual contexts.
- Model tendency to generate incorrect intermediate reasoning steps that mislead answer inference, particularly in 1B-models.
- Lindsey 2026 paper finding that models can articulate content of injected activation patterns; supports claim about self-knowledge representations
- The finding that interpretable concepts including character traits are encoded as linear directions in transformer residual streams