Context injection shifts LLM hidden states and attention, and the paper claims this drift tracks and eventually "locks in" hallucination behavior, but the reported data contain internal inconsistencies.
In: Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Shadows in the Attention: Contextual Perturbation and Representation Drift in the Dynamics of Hallucination in LLMs
Context injection shifts LLM hidden states and attention, and the paper claims this drift tracks and eventually "locks in" hallucination behavior, but the reported data contain internal inconsistencies.