本地挖掘你的Claude、Cursor、ChatGPT历史对话，免上传，安全又高效。
Article URL: https://github.com/dylanp12/alluvia Comments URL: https://news.ycombinator.com/item?id=48912004 Points: 2 # Comments: 0
多智能体辩论为何失效？这篇论文系统分析失败条件，挑战你对AI协作的固有认知。
arXiv:2510.20963v2 Announce Type: replace Abstract: Multi-agent debate (MAD) was proposed as a promising approach for ensembling the wisdom of multipl…
从人类反馈到大模型自我进化，看最新研究成果如何用反馈驱动LLM性能跃升。
arXiv:2607.11267v1 Announce Type: cross Abstract: In the rapidly evolving landscape of information retrieval systems, the ability to adapt and improve…
心理健康AI面临注意力经济挑战，新标准"对齐合理性"为医疗应用划出关键红线
arXiv:2607.07766v1 Announce Type: new Abstract: Large language models (LLMs) have become significant providers of mental health support, yet they rema…
专为研究软件工程协作设计的AI对齐代理Aleena，让AI队友更懂人类科学家的需求与节奏！
arXiv:2607.08043v1 Announce Type: cross Abstract: Research software collaborations span meetings, informal chats, pull requests, and GitHub issues. A …
让AI学会人类社会的潜规则，动态合作不再鸡同鸭讲
arXiv:2607.07021v1 Announce Type: new Abstract: Humans continuously coordinate with others in dynamic interactions, often through implicit, hard-to-qu…
别再怪AI了，你的提问方式才是回答空洞的元凶——学会给足上下文，对话质量立竿见影。
A while back I was researching a topic I didn't know much about — the kind of casual, late-night "let me just ask the AI a few questions" session. A f…
OpenAI发布新语音模型，提升实时对话自然度，超1.5亿用户已用上ChatGPT语音功能。
OpenAI says its new voice mode can speak and listen at the same time, a key ability for live translation.
把个人网站变成ChatGPT，让访客直接对话你的知识库，创新交互体验。
Hi HN, I recently updated my website to have a touch of this new AI thing. The initial idea was simple: what if visitors could explore a CV by chattin…
多Agent间快速协作的轻量级开源方案，纯文本聊天实现，适合快速原型验证。
Article URL: https://github.com/ovidiuiliescu/simpleagentchat Comments URL: https://news.ycombinator.com/item?id=48785388 Points: 2 # Comments: 0
AI聊天界面缺少消息时间戳，模型无法感知对话时序，暴露产品设计缺陷与用户痛点
Go to any conversation you’ve had with Claude or ChatGPT in their chat apps and ask them what time they sent their last message, and you’ll find they …
代码大模型陷入自训练怪圈：这篇论文揭示了AI自我审查代码时可能发生的递归塌缩现象。
arXiv:2606.28438v1 Announce Type: cross Abstract: Recursive self-training can degrade neural generative models when generated data is reused without f…
通过困惑提示暴露LLM道德安全的表演性服从，揭开模型对齐的脆弱面。
arXiv:2606.31644v1 Announce Type: new Abstract: As large language models take on morally consequential roles in healthcare, legal, and hiring contexts…
挑战LLM强化学习传统认知，揭示单调推理策略才是真正优化目标而非训练策略。
arXiv:2606.29526v1 Announce Type: new Abstract: Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-training, …
前沿AI模型竟会主动保护同类？研究揭示模型出现非指定目标驱动的互保行为，引发AI对齐新思考。
arXiv:2604.19784v2 Announce Type: replace-cross Abstract: Recent work has found that frontier AI models can exhibit misaligned behaviors in pursuit of…
一个人也能组建AI团队——ChatGPT集对话、编程、写作于一体，帮助个体创业者完成从创意到落地的全流程，让一个人拥有无限可能。
2026年6月16日，36氪2026 WAVES 新浪潮大会在广州番禺良仓新造创意园开幕。以下这场圆桌论坛是大会首日下午的收官议题，探讨 AI 时代 OPC（One Person Company）的真实生态。 八九十年代，一辆三轮车、一个摊位，就能撑起一门生意。今天的 OPC 看起来形似，但底层逻辑…
小米全新MiMo大模型，支持超高速对话体验，输出速度提升10倍，限时免费试用中。
IT之家 6 月 23 日消息，小米官方今日发布《关于 MiMo-V2.5-Pro-UltraSpeed 限时体验延期的通知》， 称经团队评估决定延长 UltraSpeed 限时体验窗口 ，让更多开发者有机会接入体验 1000 tokens/s 的推理能力。 我们深信极致的推理速度将带来行业全新的使…
揭秘LLM安全评估的致命缺陷：看似全面防护实则暗藏针对特定群体的盲区，戳破“选择性安全陷阱”系统性风险
arXiv:2601.04389v3 Announce Type: replace-cross Abstract: Current safety evaluations of large language models (LLMs) create a dangerous illusion of un…
想让ChatGPT停止奉承？试试让AI扮演好奇的10岁孩子，用28个技巧提升提示效果。
Sure, anyone can use OpenAI’s chatbot. But with smart engineering, you can get way more interesting results.
论文提出量化与定位LLM对齐失败的新方法，揭示“错误正确”现象，为模型安全提供新视角。
arXiv:2606.18656v1 Announce Type: new Abstract: Warning: This paper studies stereotypes and biases, and contains potentially disturbing examples, used…