TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

A Survey of AI Agent Protocols

91 点作者 distalx20 天前

6 条评论

klabb319 天前
It’s good to see a new fresh push for protocols and interop. However, I don’t think it will do the thing we hope for, for non-technical reasons.<p>During Web 2.0, we saw similar enthusiasm. Instead of AI agents or blockchain, every modern company had an API exposed. For instance, Gmail- and Facebook chat was usable with 3p client apps.<p>What killed this was not tech, but business. The product wasn’t say social media, it was ad delivery. And using APIs was considered a bypass of funnels that they want to control. Today, if you go to a consumer service website, you will generally be met with a login&#x2F;app wall. Even companies that charge money directly (say 23&amp;me ad an egregious example) are also data hoarders. Apple is probably a better example. There’s no escape.<p>The point is, protocols is the easy part. If the economics and incentives are the same as yesterday, we will see similar outcomes. Today, the consumer web is adversarial between provider ”platforms”, ad delivery, content creators, and the products themselves (ie the people who use them).
评论 #43895106 未加载
simonw20 天前
I scrolled straight to section 2.1 &quot;Definition and Characteristics of LLM agents&quot; to find out which of the many definitions of &quot;agent&quot; they are using here.<p>They went for LLM + short-term and long-term memory + planning + tool using + action execution.<p>Presumably &quot;planning&quot; here is covered by any LLM that can do &quot;think step by step&quot; reasonably well?<p>It wasn&#x27;t clear to me what the difference between &quot;tool using&quot; and &quot;action execution&quot; was.<p>I haven&#x27;t seen a definition that specifically encompasses both short- and long-term memory before. They say:<p>&gt; This dual memory system allows agents to maintain conversation continuity while building knowledge over time.<p>So presumably, this is the standard LLM chat conversation log plus a tool that can decide to stash extra information in a permanent store - similar to how ChatGPT&#x27;s memory feature worked up until about four weeks ago.
评论 #43884679 未加载
评论 #43885489 未加载
评论 #43884838 未加载
评论 #43884978 未加载
revskill20 天前
Is AI Agent just LLM wrapper ? Is there anything more interesting to it ?
评论 #43884815 未加载
评论 #43885529 未加载
评论 #43887030 未加载
评论 #43886036 未加载
bn-l20 天前
This is a very weird paper. They go through all these protocols yet haven’t don’t mention the most promising one (smolagents): <a href="https:&#x2F;&#x2F;github.com&#x2F;huggingface&#x2F;smolagents">https:&#x2F;&#x2F;github.com&#x2F;huggingface&#x2F;smolagents</a>
评论 #43886612 未加载
anonymousDan20 天前
The hype around agent protocols reminds me of the emperor&#x27;s new clothes. There&#x27;s just nothing to it from a technical perspective.
评论 #43884742 未加载
评论 #43886637 未加载
评论 #43889013 未加载
评论 #43884952 未加载
评论 #43884610 未加载
评论 #43885289 未加载
评论 #43885343 未加载
评论 #43885035 未加载
评论 #43885162 未加载
评论 #43884879 未加载
sgt10120 天前
Doesn&#x27;t even mention KQML or FIPA - staggeringly ignorant.
评论 #43888923 未加载
评论 #43886802 未加载
评论 #43886318 未加载