新闻
Tokenizer Expansion: Upgrading a Model's Tokenizer in Place
Liquid AI · 发布于 · 阅读约3分钟
30秒读懂
- 发生了什么
- Liquid AI expanded LFM2.5-8B-A1B's tokenizer from 65K to 128K vocabulary in place, without retraining, improving compression for underrepresented languages.
- 为何重要
- Matters for engineers building on-device models for Hindi, Vietnamese, Thai, and Bengali speakers where token count directly impacts latency and energy consumption.
- 注意
- The larger vocabulary slows per-token decoding by 7-10%, and the method only works when you control the tokenizer and have its original merge rules available.
收听本摘要
- tokenizer
- token
The Agent Architect
每周一个模式、一个权衡、一个生产事故案例。为构建智能体系统的人准备的每周简报。
每周一封邮件,一键退订。您的地址仅用于发送简报。