跳到正文

主题

research

当前情况

编辑部还没有为这个话题留下带日期的说明。

New benchmark: no coding agent reliably builds the agent you asked for

AgentSWE tests whether coding agents can create, edit and optimise other AI agents from a plain-language requirement. None came close, the authors report.

主题是报道上已有的标签。至少两篇报道使用时,才会列在这里。

  1. Wooden letter tiles spelling STRAWBERRY on a table, a magnifying glass over the letter R, warm light. Generated photograph.
    Made with AI
  2. A researcher pointing at a large screen showing a neat grid of glowing activation tiles, dark lab. Generated photograph.
    Made with AI
  3. A tall stack of printed research papers with red pen marks and sticky notes on a university desk, lamp light. Generated…
    Made with AI
  4. 三个人在大学公共休息室里核对一摞很高的论文。生成照片。并非本文提及的任何人。
    Made with AI

问题

这一页有什么?

这一页列出 5 篇标有 research 的报道。

当前情况是什么?

编辑部还没有为这个话题留下带日期的说明。

来源在哪里?

每篇报道都链向它使用的来源。

research · The AI Pin