2 papers
cs.CL2025
Measuring the Robustness of Reference-Free Dialogue Evaluation Systems
Justin Vasselli, Adam Nohejl, Taro Watanabe
Advancements in dialogue systems powered by large language models (LLMs) have outpaced the development of reliable evaluation metrics, particularly for diverse and creative respons…
cs.CL2025
Dispersion Measures as Predictors of Lexical Decision Time, Word Familiarity, and Lexical Complexity
Adam Nohejl, Taro Watanabe
Various measures of dispersion have been proposed to paint a fuller picture of a word's distribution in a corpus, but only little has been done to validate them externally. We eval…