[2302.04761] Toolformer: Language Models Can Teach Themselves to Use Tools
5
公开标注数
5
参与人数
2026-07-19 09:42:26
首次 Whisper
本页的公开 Whisper
划选高亮2026-07-19 13:03:26
原文高亮摘录
“They also, paradoxically, struggle with basic functionality, such as a”
Whisper 随想笔记
I mean, isn't that why we have calculators? LMs just need tools, makes sense.
划选高亮2026-07-19 12:54:26
原文高亮摘录
“They also, paradoxically, struggle with basic functionality, such as a”
Whisper 随想笔记
Haha, so true — they can write essays but can't do simple math.
划选高亮2026-07-19 10:00:26
原文高亮摘录
“Language models (LMs) exhibit remarkable abilities to solve new tasks from just a few examples or textual instructions, especially at scale.”
Whisper 随想笔记
So the trick is just letting them call APIs? Kinda obvious in hindsight.
划选高亮2026-07-19 09:51:26
原文高亮摘录
“Language models (LMs) exhibit remarkable abilities to solve new tasks from just a few examples or textual instructions, especially at scale.”
Whisper 随想笔记
I've seen this paradox firsthand—models that ace essays but fail 2+2.
划选高亮2026-07-19 09:42:26
原文高亮摘录
“Language models (LMs) exhibit remarkable abilities to solve new tasks from just a few examples or textual instructions, especially at scale.”
Whisper 随想笔记
Scaling really does magic for few-shot learning, but yeah, basic math still trips them up.
分享本页 Whisper
短链接
https://domwhisper.com/s/200002328137嵌入代码
<iframe src="https://domwhisper.com/embed/200002328137" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>