top-p-sampling-966034d4·1 events·first seen Aliases: Top-p sampling, Top-k sampling
This paper introduces the Word Coverage Score (WCS), a metric that quantifies how much contextually appropriate low-frequency vocabulary is pruned away by standard sampling strategies (Top-p, Top-k, Min-p) in LLMs. The authors audit open-weight models against human-authored corpora to measure the 'lexical survival rate' of high-information words under typical decoding defaults. Their findings provide quantitative evidence that industry-standard sampling parameters act as unintended censorship mechanisms, suppressing linguistic diversity even when rare words exist within the model's probability distribution. The WCS is proposed as a diagnostic tool for tuning the coherence–lexical-richness trade-off.