Stylistic collapse in LLMs is largely pre-alignment and scale-independent; strong entropy regularization reduces it far more than weak smoothing or RLHF.
Fraser, Hillary Dawkins, and Svetlana Kiritchenko
1 Pith paper cite this work, alongside 40 external citations. Polarity classification is still indexing.
1
Pith paper citing it
40
external citations · OpenAlex
fields
cs.CL 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
From Context Shift to Stylistic Collapse: Why Training Objectives Matter More Than Scale
Stylistic collapse in LLMs is largely pre-alignment and scale-independent; strong entropy regularization reduces it far more than weak smoothing or RLHF.