[1810.04805] BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
5
公开标注数
5
参与人数
2026-07-15 09:42:01
首次 Whisper
本页的公开 Whisper
划选高亮2026-07-15 13:03:01
原文高亮摘录
“BERT is designed to pre-train deep bidirectional representations”
Whisper 随想笔记
ok but does bidirectional really help that much or is it just the bigger model?
划选高亮2026-07-15 12:54:01
原文高亮摘录
“BERT is designed to pre-train deep bidirectional representations”
Whisper 随想笔记
finally someone actually reading both directions, not just left-to-right
划选高亮2026-07-15 10:00:01
原文高亮摘录
“BERT, which stands for Bidirectional Encoder Representations from Transformers”
Whisper 随想笔记
This acronym is a pain to explain to my boss every single time.
划选高亮2026-07-15 09:51:01
原文高亮摘录
“BERT, which stands for Bidirectional Encoder Representations from Transformers”
Whisper 随想笔记
I still don't get why bidirectional is such a big deal, seems obvious.
划选高亮2026-07-15 09:42:01
原文高亮摘录
“BERT, which stands for Bidirectional Encoder Representations from Transformers”
Whisper 随想笔记
Finally a model that actually reads both ways, not just left to right.
分享本页 Whisper
短链接
https://domwhisper.com/s/32f3c7d144f0嵌入代码
<iframe src="https://domwhisper.com/embed/32f3c7d144f0" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>