Line of inquiry
Inquiring lines›How do training choices shape mode…›How do systems prioritize structur…›this line of inquiry
Why are hallucinations robust to training-based intervention strategies?
A broader line of inquiry — a family of 11 specific questions the research asks around this. Follow one into its inquiring-line page, or move sideways to a related line below.
Questions in this line of inquiry 11
Specific inquiring lines the field asks around this — ordered from the most general framing down to the most specific angle.
- Can filtering unknown examples during fine-tuning prevent hallucination increases?
- Why does model confidence fail to detect hallucinations on rare entity pairs?
- Does cross-example gradient contamination explain finetuning-induced hallucination patterns?
- Why does model confidence fail to detect hallucinations about rare entities?
- Can scaling up contradictory training data overcome unpredictable override effects?
- Can pretraining-frequency signals alone prevent RAG systems from confabulating about common knowledge?
- Why do interventions for hallucination or automation bias fail to address capability misattribution?
- Why does test accuracy improve after training accuracy reaches 100 percent?
- Can held-out validation gates prevent optimizer hallucinations in skill proposals?
- Can training data analysis predict which samples will cause unintended personality changes?
- How does modified PPO handle samples from much older model versions?