Technology Radar
Published : Apr 02, 2025
NOT ON THE CURRENT EDITION
This blip is not on the current edition of the Radar. If it was on one of the last few editions, it is likely that it is still relevant. If the blip is older, it might no longer be relevant and our assessment might be different today. Unfortunately, we simply don't have the bandwidth to continuously review blips from previous editions of the Radar.
Understand more
Apr 2025
Assess
BERT（Bidirectional Encoder Representations from Transformers）的继任者 ModernBERT 是一系列新一代的 encoder-only transformer 模型，专为广泛的自然语言处理（NLP）任务设计。作为一个可直接替代 BERT 的升级版本，ModernBERT 不仅提升了性能和准确性，还解决了 BERT 的一些局限——特别是通过引入“交替注意力”（Alternating Attention）实现了对极长上下文长度的支持。 对于有 NLP 需求的团队，在默认选择 通用生成式模型 之前，请优先考虑 ModernBERT。
