Diversity collapse in LLMs arises from order and shape miscalibration in token probability distributions at inference time, not from sampling methods.
Benchmarking linguistic diversity of large language models.Transactions of the Association for Computational Linguistics, 13:1507–1526
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Sampling More, Getting Less: Calibration is the Diversity Bottleneck in LLMs
Diversity collapse in LLMs arises from order and shape miscalibration in token probability distributions at inference time, not from sampling methods.