<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[AI 客服回答太像模板，是提示词问题还是产品问题]]></title><description><![CDATA[<p dir="auto">我们客服机器人被吐槽“像复制粘贴”。我改了几版提示词，要求自然一点，效果还是很尬。</p>
]]></description><link>https://localaihub.com/topic/140/ai-客服回答太像模板-是提示词问题还是产品问题</link><generator>RSS for Node</generator><lastBuildDate>Wed, 03 Jun 2026 19:29:16 GMT</lastBuildDate><atom:link href="https://localaihub.com/topic/140.rss" rel="self" type="application/rss+xml"/><pubDate>Sat, 09 May 2026 18:04:00 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 17:26:00 GMT]]></title><description><![CDATA[<p dir="auto">我先把差评样本拉出来，分类型调。之前确实全靠感觉。</p>
]]></description><link>https://localaihub.com/post/1386</link><guid isPermaLink="true">https://localaihub.com/post/1386</guid><dc:creator><![CDATA[小曹]]></dc:creator><pubDate>Sun, 10 May 2026 17:26:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 16:06:00 GMT]]></title><description><![CDATA[<p dir="auto">对，AI 客服不是作文比赛。产品动作能解决的，不要都交给模型说。</p>
]]></description><link>https://localaihub.com/post/1385</link><guid isPermaLink="true">https://localaihub.com/post/1385</guid><dc:creator><![CDATA[阿白]]></dc:creator><pubDate>Sun, 10 May 2026 16:06:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 13:42:00 GMT]]></title><description><![CDATA[<p dir="auto">还有按钮。能点“转人工”“查看订单”“重新描述问题”，就不要让模型用 200 字解释。</p>
]]></description><link>https://localaihub.com/post/1384</link><guid isPermaLink="true">https://localaihub.com/post/1384</guid><dc:creator><![CDATA[leaf_1997]]></dc:creator><pubDate>Sun, 10 May 2026 13:42:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 12:11:00 GMT]]></title><description><![CDATA[<p dir="auto">这个很具体，我试试。</p>
]]></description><link>https://localaihub.com/post/1383</link><guid isPermaLink="true">https://localaihub.com/post/1383</guid><dc:creator><![CDATA[小曹]]></dc:creator><pubDate>Sun, 10 May 2026 12:11:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 09:05:00 GMT]]></title><description><![CDATA[<p dir="auto">可以设置一句硬规则：“不在第一句道歉，除非用户明确表达损失或投诉。”很多废话从这里来的。</p>
]]></description><link>https://localaihub.com/post/1382</link><guid isPermaLink="true">https://localaihub.com/post/1382</guid><dc:creator><![CDATA[Grace]]></dc:creator><pubDate>Sun, 10 May 2026 09:05:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 08:39:00 GMT]]></title><description><![CDATA[<p dir="auto">我这边做过一个小改动：回答先给结论，再给步骤，最后再说注意事项。投诉少了一些。</p>
]]></description><link>https://localaihub.com/post/1381</link><guid isPermaLink="true">https://localaihub.com/post/1381</guid><dc:creator><![CDATA[半糖]]></dc:creator><pubDate>Sun, 10 May 2026 08:39:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 06:59:00 GMT]]></title><description><![CDATA[<p dir="auto">还有知识库答案本身可能就像模板。模型只是把原来的模板换个说法。</p>
]]></description><link>https://localaihub.com/post/1380</link><guid isPermaLink="true">https://localaihub.com/post/1380</guid><dc:creator><![CDATA[小潘同学]]></dc:creator><pubDate>Sun, 10 May 2026 06:59:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 06:46:00 GMT]]></title><description><![CDATA[<p dir="auto">那就先让他们补样本。没有失败样例，提示词改到最后就是玄学。</p>
]]></description><link>https://localaihub.com/post/1379</link><guid isPermaLink="true">https://localaihub.com/post/1379</guid><dc:creator><![CDATA[陈一]]></dc:creator><pubDate>Sun, 10 May 2026 06:46:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 04:22:00 GMT]]></title><description><![CDATA[<p dir="auto">业务方现在只给了几条截图，没有系统导出。</p>
]]></description><link>https://localaihub.com/post/1378</link><guid isPermaLink="true">https://localaihub.com/post/1378</guid><dc:creator><![CDATA[小曹]]></dc:creator><pubDate>Sun, 10 May 2026 04:22:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 01:53:00 GMT]]></title><description><![CDATA[<p dir="auto">建议先拿 50 条差评对话标注：废话多、没命中问题、权限不敢答、流程太长、要人工。别混在一起调。</p>
]]></description><link>https://localaihub.com/post/1377</link><guid isPermaLink="true">https://localaihub.com/post/1377</guid><dc:creator><![CDATA[melo]]></dc:creator><pubDate>Sun, 10 May 2026 01:53:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sun, 10 May 2026 01:44:00 GMT]]></title><description><![CDATA[<p dir="auto">我见过很多机器人坏在“安抚语”太多。业务方觉得亲切，用户觉得废话。</p>
]]></description><link>https://localaihub.com/post/1376</link><guid isPermaLink="true">https://localaihub.com/post/1376</guid><dc:creator><![CDATA[Tina]]></dc:creator><pubDate>Sun, 10 May 2026 01:44:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sat, 09 May 2026 23:35:00 GMT]]></title><description><![CDATA[<p dir="auto">这种话删掉一半就好了。客服不是越礼貌越好，尤其技术支持场景，用户想要的是下一步。</p>
]]></description><link>https://localaihub.com/post/1375</link><guid isPermaLink="true">https://localaihub.com/post/1375</guid><dc:creator><![CDATA[何必呢]]></dc:creator><pubDate>Sat, 09 May 2026 23:35:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sat, 09 May 2026 21:15:00 GMT]]></title><description><![CDATA[<p dir="auto">都有。它会先说“您好，非常理解您的心情”，然后绕半天。</p>
]]></description><link>https://localaihub.com/post/1374</link><guid isPermaLink="true">https://localaihub.com/post/1374</guid><dc:creator><![CDATA[小曹]]></dc:creator><pubDate>Sat, 09 May 2026 21:15:00 GMT</pubDate></item><item><title><![CDATA[Reply to AI 客服回答太像模板，是提示词问题还是产品问题 on Sat, 09 May 2026 18:27:00 GMT]]></title><description><![CDATA[<p dir="auto">先别急着改语气。用户说像模板，是因为每句话都像模板，还是因为没解决问题？</p>
]]></description><link>https://localaihub.com/post/1373</link><guid isPermaLink="true">https://localaihub.com/post/1373</guid><dc:creator><![CDATA[小蓝]]></dc:creator><pubDate>Sat, 09 May 2026 18:27:00 GMT</pubDate></item></channel></rss>