AI 模型中的“上下文投毒”概念已浮出水面，即对话早期引入的错误信息可能导致模型生成有缺陷的响应。这一现象在一个评论串中被讨论，强调了模型如果未得到妥善纠正，可能会保留并传播错误。另外，游戏 Marvel Tōkon: Fighting Souls 因其 PC 版移植问题（包括 PlayStation 账户要求和反作弊问题）在 Steam 上获得了褒贬不一的评价，开发商 Arc System Works 已对此作出回应。 AI
排序理由 该集群讨论了一个小众的 AI 概念和一个电子游戏评论，两者均未达到重大新闻的门槛。
在 Mastodon — fosstodon.org 阅读 →
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
🤖 Learned the term "context poisoning" today and now I can't stop noticing it Someone explained this to me in a comment thread and it's been rattling around in my head since. The idea: in a long conversation, if the model says something wrong and you correct it, that correct... 📰…
🎮 Marvel Tokon Stumbles to Mixed Steam Reviews as 'Miserable' PC Port Complaints Prompt Response From Arc System Works Marvel Tokon: Fighting Souls stumbled on Steam with "Mixed" user reviews as players complain of PlayStation account requirements, anti-cheat roadblocks, and othe…