一作ICML 2026Poster· Deep Learning / Large Language Models
Fast Byte Latent Transformer
Julie Kallini · Artidoro Pagnoni · Tomasz Limisiewicz · Gargi Ghosh · Luke Zettlemoyer · Christopher Potts · Xiaochuang Han · Srinivasan Iyer
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Julie Kallini · Artidoro Pagnoni · Tomasz Limisiewicz · Gargi Ghosh · Luke Zettlemoyer · Christopher Potts · Xiaochuang Han · Srinivasan Iyer
Milan Miletić · Julie Kallini · Ekaterina Shutova
Multilingual language models often exhibit performance disparities across languages that can arise as early as the tokenization stage. Widely-used subword tokenization approaches favor high-resource languages, and tokenizer-free methods still yield longer sequences for scripts with a higher bytes...