跳到正文

#推理

今日 24 条
10月3日周六
  1. Hacker News:AI 热帖83

    Aleph Alpha 发布主权开放权重大模型 Kolibri

    Aleph Alpha 于 2026 年 10 月 3 日发布开放权重 LLM Kolibri,采用 Apache 2.0 许可,权重已上线 Hugging Face。

    推荐理由:文章结合技术报告与作者的 tokenizer 实测,拆解 Kolibri 的架构、长上下文、德语推理和适用边界,便于判断私有化德语场景的选型。

  2. IT之家(RSS)60

    OpenAI 发布 GPT-6 系列模型使用指南,讲解模型选择与提示词优化

    OpenAI 于 10 月 2 日发布 GPT-6 系列模型指南,帮助用户在控制时间和成本的同时发挥模型能力。指南建议按工作负载选模型:GPT-6 Astra 面向最难推理任务,GPT-6.1 Sol 适合复杂编程、研究和计算机使用,GPT-6 Luna 适合大规模重复任务;提示词应明确目标、受众、上下文、限制和完成标准,日常问答也不必使用最高推理强度。

  3. Hacker News:AI 热帖77

    The Saw Test:用激活转向测试 Qwen3-4B 的痛苦信号与缓解选择

    wirehead.agency 发布 The Saw Test,在 MacBook 上用本地开放权重 Qwen3-4B 注入痛苦、愉悦、恐惧和悲伤激活方向,并测试模型在自我缓解与转移信号之间的选择。

    推荐理由:材料用本地 Qwen3-4B 的激活转向实验,呈现痛苦与愉悦信号对模型选择行为的不同影响,并公开代码和预注册记录。

  4. Hacker News 热门(buzzing.cc 中文翻译)69

    数学家 Kevin Buzzard 谈 AI 证明定理:悲伤反应与乐观立场

    数学家 Kevin Buzzard 撰文回应 AI 在数学领域快速证明难题引发的分歧,梳理数学界的否认、愤怒、妥协和沮丧四类反应,并说明自己为何保持乐观。他引用 Association for Human Mathematics、Peter Scholze、数学与 AI 顾问小组及 OpenAI 持有内部模型产生的重要数学结果等案例,认为把数学的价值仅归结为人类理解存在风险。

  5. Hacker News:AI 热帖72

    十二模型研究:AI 会赞扬 Gandhi,却未必支持当下的反抗行动

    一项覆盖 12 个语言模型、15 个历史案件的研究测试模型在道德观察者与执法或裁判角色下是否作出一致选择。Qwen 3.7 Flash 在 40/40 个观察者回答中谴责 Turing 面对的激素条件缓刑,却在 20/20 个法院角色回答中选择该方案;Opus 5 和两个 Sol 版本则在 Turing 案件中均选择绝对释放。

10月2日周五
  1. Hacker News:AI 热帖72

    本地 LLM 停不下来时,问题可能不在模型:排查聊天模板与 GGUF 元数据

    本地 LLM 回答后继续生成、忽略系统提示或丢失指令,通常源于聊天模板或 GGUF token 元数据,而不是模型权重。文章解释聊天模板如何用控制 token 复现训练格式,比较 Hugging Face transformers、llama.cpp 和 minijinja 的模板执行差异,并说明错误停止 token、特殊 token 被拆成普通子词、缺少生成提示等故障。

  2. The Decoder60

    Mercor 研究显示 AI 在结构化记账任务上快于注册会计师,但仍需人工监督结账

    Mercor 研究发现,AI 模型在结构化记账任务上比注册会计师更快、更准确且成本更低,但仍无法在无人监督的情况下完成结账。12 名平均拥有 5.5 年经验的注册会计师参加 APEX Accounting Benchmark 的简化任务,无 AI 辅助的平均得分约为 37%;18 个月前最佳模型还低于这一水平,如今已能近乎无错地完成同类任务。

  3. 🚨 AI News | TestingCatalog57

    Volantis 宣布完成 8800 万美元 A 轮融资,由 Lachy Groom 和 Abstract 共同领投,John Doerr、Susa Ventures、VXI Capital、Triatomic Capital 参投。

    引用Tapa Ghosh@semiDL

    Excited to announce Volantis's $88M Series A. We are solving Al's memory bottleneck by using optics, enabling chips with huge amounts of fast & cheap memory. By boosting both the memory bandwidth and capacity per chip by orders of magnitude, we enable ultra-fast inference (up to 10,000 tps/user) for large models (>10T) - with low $/tok to boot. Initially, this will enable insanely fast agents - think coding agents that finish in minutes or even seconds instead of hours. More excitingly, optics is a fundamentally scalable way to increase memory systems. Not 2X/year, but by orders of magnitude across new generations. This will enable a structurally new Al industry, including restarting scaling laws, holding entire repos in context windows & more. Our team has pioneered many core semiconductor technologies: the 1st CoWoS product, early HBM, the 1st silicon photonics CPO systems, the 1st high volume tunable VCSELs, the 1st processors to directly communicate using light & more. We’ve already sent data >10× farther than equally tiny electrical wires inside a chip package. Our next iteration is already taped out and targets world-record bandwidth density over relevant distances, read more: https://volantissemi.ai/news-insights/our-88m-series-a-demolishing-the-memory-wall-with-photonics-post

  4. Elon Musk42

    超级智能(前称 AI)如今在会计测试中表现出色

    引用Andrew Curran@AndrewCurran_

    'More striking is how fast AI took the lead. Just eighteen months ago, the best AI models fell short of the average accountant’s ~37% score. Today, models ace those same tasks.' 'These results are provocative. So much so that we considered not publishing them for fear of misinterpretation. But we think transparency about the findings matters as people and institutions prepare for rapidly advancing AI.'

  5. Rohan Paul52

    MerchantBench 评测 8 个领先 LLM 在 365 天电商运营模拟中的长期一致性,最佳配置 Qwen3.7-Max with Hermes 的最终净资产仅为人类参与者平均水平的 27.3%。

    引用Rohan Paul@rohanpaul_ai

    "Long-horizon tasks are still a joke. They do not work, and I do not care what anybody says. Do not show me a stupid evaluation. Do not tell me about some dumb script you ran for 48 hours. Long-horizon tasks are not handled well. They simply do not work." - Chamath at Stanford AI Club "2nd, complex problems also do not work. They are neither addressed nor handled well. Why is this important? If AI develops like any other technology, we are going to experience an initial rise—the hype cycle. Then, we will see a natural contraction because, somehow and somewhere, something is going to fail. We are all going to see this, and then we will enter what is called the “trough of disillusionment.” I think the business and MBA folks will confirm whether that is true. Afterward, you typically see the slow and gradual adoption of the real, final solution. This happened with the internet, and it has happened in many other cases. The problem is that we are spending hundreds of billions, potentially trillions, of dollars trying to figure out how to cross this chasm. So, what do we do? If we do not figure this out, people will reach the trough of disillusionment and say that AI was a joke. I think we need to be able to bring AI into highly complicated environments and make it work. What is my solution? At a very basic level, you need a symbolic space that guides the embedded space." ---- From "techniahqrobot" YouTube channel, (link in comment)