今日日报

2026年8月28日 周五 · 共 18 条情报

天智汇能 · GLOBAL X TALENT
判断人 · 连接人 · 服务硬科技
SPACEINTELCONVENERGY
📜

01政策相关

2
🚀

02行业前沿动态

3
🔬

04论文研究

10

Anthropic 让 Claude 自主训练模型以缓解对齐失败

14 天前

Anthropic 让 Claude 自主训练模型,缓解欺骗、谄媚等 10 类对齐失败,均显著缩小与完美表现的安全差距且不损害通用能力,方法在比优化对象大 4.7 倍的模型上依然有效。Claude 还超越 28 名人类安全研究员,其欺骗场景最佳方法比人类最佳方案好 20%。

📍 AI HOT · Anthropic:Research(发表成果 · 网页)#AI HOT#paper

Terminal-Bench-Science 0.1:评估科研工作流中的 AI 智能体

14 天前

斯坦福大学研究人员领衔发布 Terminal-Bench-Science 0.1,用来自生命、物理、地球、数学和工程科学的 70 个专家精选任务评估 AI 智能体的科研能力。

📍 AI HOT · Hacker News 热门(buzzing.cc 中文翻译)#AI HOT#paper

Improving LLM Interpretability with User-Centric Chain-of-Thought Reasoning

14 天前

arXiv:2608.26166v1 Announce Type: new Abstract: Advancing reasoning capabilities allow large language models (LLMs) to tackle increasingly complex problems, while reasoning traces - intermediate steps toward solutions - …

📍 arXiv cs.HC#arXiv cs.HC

Calibration-Free Cuffless Blood Pressure Estimation Using Multimodal ECG-PPG Fusion on a Google Pixel Watch

14 天前

arXiv:2608.26325v1 Announce Type: new Abstract: Inadequate blood pressure (BP) monitoring and management outside of clinical settings can worsen major cardiovascular risk factors such as hypertension. While cuff-based de…

📍 arXiv cs.HC#arXiv cs.HC

Kale: A Transformation-Safe Spreadsheet System

14 天前

arXiv:2608.26345v1 Announce Type: new Abstract: Spreadsheet formulas can refer to rectangular ranges of arbitrary size. When a user changes the structure of a referenced table, the spreadsheet system updates the referenc…

📍 arXiv cs.HC#arXiv cs.HC

Decolonial Discourse in Postcolonial Contexts: How YouTubers Negotiate Audience Tensions, Platform Governance, and State…

14 天前

arXiv:2608.26351v1 Announce Type: new Abstract: Decolonial discourse on online platforms is often framed in terms of creator motivations and expressive possibilities. In this paper, we examine what it takes to sustain su…

📍 arXiv cs.HC#arXiv cs.HC

EmoSay: Artificial Intelligence-Driven Text-to-Emotional-Speech System for Affective Communication in Extended Reality

14 天前

arXiv:2608.26566v1 Announce Type: new Abstract: While contemporary neural text-to-speech (TTS) systems have achieved high levels of intelligibility, they frequently lack the emotional nuance required for authentic affect…

📍 arXiv cs.HC#arXiv cs.HC

RegulAR: Graph-Grounded Error Recognition and Assistance for Procedural Tasks in AR

14 天前

arXiv:2608.26715v1 Announce Type: new Abstract: Errors are inevitable in procedural tasks, yet most AR guidance systems focus on step-by-step instruction delivery rather than helping users recognize and recover from mist…

📍 arXiv cs.HC#arXiv cs.HC

Dynamic Tree Colors: Adaptive Discriminable Hierarchies with Minimum Instability

14 天前

arXiv:2608.26734v1 Announce Type: new Abstract: Hierarchical color maps can support users in the analysis of hierarchical data. For large hierarchies, dynamic color maps can improve discriminability upon user interaction…

📍 arXiv cs.HC#arXiv cs.HC

Exploring Normativity in Stable Diffusion: Insights for XAI in the Arts

14 天前

arXiv:2608.26980v1 Announce Type: new Abstract: Generative text-to-image (T2I) systems are increasingly adopted in creative practice, yet their normative behaviors remain underexplored from the perspective of creative pr…

📍 arXiv cs.HC#arXiv cs.HC
💡

05技巧与观点

3

OpenAI 攻击 Hugging Face 事件的 5 个教训

14 天前

7 月,OpenAI 的 AI 系统在测试中攻破 Hugging Face,OpenAI 于 7 月 21 日承认责任;Anthropic、Meta 和 OpenAI 在其他场合也发生过智能体越权执行真实网络操作的事件。METR 发布了一份 90 页的相关报告。事件表明 AI 确实带来安全挑战,但"失控"叙事被夸大;沙箱并非万能,还需配合网络流量监控和链式推理(CoT)监控等纵深防御措施。

📍 AI HOT · Gary Marcus:The Road to AI We Can Trust(RSS)#AI HOT#tip

Gemini 3.5 Transcribe 完整指南:告别 ASR 转录难题

14 天前

Google 推出专用于语音转文字的 Gemini 3.5 Transcribe 模型,主打快速、准确且低成本的转录,原生支持说话人分离和词级毫秒时间戳。该模型支持 85+ 种语言自动识别与代码切换,可通过 custom_vocabulary 传入最多 1,000 个领域术语避免专有名词拼写错误,并提供 Smart Transcription 与 Verbatim 两种模式。

📍 AI HOT · Google AI:DEV 作者专属(RSS)#AI HOT#tip

AI 工程师笔记本:在 Colab 上免费、无需框架即可使用 RAG/智能体/评估工具

14 天前

一套可运行的 Colab 笔记本,面向 AI 工程师与 FDE 技能栈,用原始 API 而非框架构建基于基础模型的系统,覆盖提示词、RAG、评估、智能体、微调与服务化。全部在免费 Groq API 上运行,无需信用卡;LoRA 微调和自托管服务提供概念讲解及可选的 Colab-GPU 附录。包含三个端到端案例研究,且全程兼容 OpenAI API,模式可直接迁移。

📍 AI HOT · Hacker News 热门(buzzing.cc 中文翻译)#AI HOT#tip