AI
AI 资讯
alishangtian.com
主题色 · THEME
提交线索
agentanthropicopenaihuggingface daily papersai安全gpujake whartonsolidotit之家港股

搜索：Deep

共命中 50 条（服务端检索）
从搜索框到 Deep Research：Agentic 检索怎么把一次查询变成一场调查
Google 比 OpenAI 早七周发布 Deep Research，但把「研究计划让用户改批」产品化的也是它——agentic 检索的通用循环是：规划、迭代检索、阅读、反思补漏、交叉验证、带引用报告。本文拆解这个循环的两种实现路线（端到端 RL vs 显式编排），用 BrowseComp 上「裸模型不足 10% vs deep research 51.5%」的差距说明多步浏览行为本身值多少分，也把「慢不等于对」的引用可靠性研究摆上台面。
智能体 精选 · 原创 · 10-8 阅读 204·访客 199
Beyond Watching and Searching: Benchmarking Agents for Video-centered Deep Research Reports
Deep research agents synthesize open-web evidence into long reports, but current benchmarks do not evaluate research cen…
智能体 HuggingFace Daily Papers · 10-22 阅读 26·访客 26
Deep Search in Real-World Web Environments
Despite substantial progress in Deep Search agents, their ability to conduct end-to-end research in real-world open Web …
智能体 HuggingFace Daily Papers · 10-19 阅读 26·访客 26
OneSearch-VL: Unified Multimodal Deep Research Agent for Image and Video
Single-image, multi-image, and video deep research require different visual operations but share a workflow of visual gr…
智能体 HuggingFace Daily Papers · 10-8 阅读 50·访客 49
Incremental Open-Ended Deep Research with Structured Harness
Existing Open-Ended Deep Research (OEDR) systems primarily generate reports from scratch, making them inefficient for sc…
行业动态 HuggingFace Daily Papers · 10-8 阅读 21·访客 21
IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis
Deep search requires LLM agents to decompose complex queries, search for evidence, and synthesize grounded answers, yet …
智能体 HuggingFace Daily Papers · 9-24 阅读 54·访客 53
Deep Fusion Memory: A Chunk-Level Parametric Memory for Large Language Models
Large language models need effective ways to acquire and use external knowledge beyond what they retain from pretraining…
行业动态 HuggingFace Daily Papers · 10-18 阅读 9·访客 9
OmniCapBench: A Deep-Structured Evaluation Framework for Fine-Grained Audio-Visual Captioning
Multimodal large language models (MLLMs) are rapidly evolving toward continuous audio--visual reasoning, creating an urg…
研究前沿 HuggingFace Daily Papers · 10-8 阅读 22·访客 22
Exa Launches Agent Ultra: A Subagent Swarm Deep Research API Built for Exhaustive List Building
Exa has released **Agent Ultra**, the highest effort level of its Exa Agent API. It is built for research that must run …
智能体 MarkTechPost · 9-26 阅读 74·访客 73
Sakana AI hires Jürgen Schmidhuber, inventor of deep learning, world models, and your next ChatGPT update
Sep 24, 2026 Silicon Valley's next big AI idea is probably already sitting in a 1991 paper by Sakana AI's new chief advi…
研究前沿 The Decoder · 9-25 阅读 41·访客 41
Deep Persona: A Psychologically Grounded Architecture and Evaluation Framework for Role-Playing Agents and Simulations
Existing approaches to persona simulation with Large Language Models (LLMs) mostly rely on shallow character description…
智能体 HuggingFace Daily Papers · 9-6 阅读 48·访客 47
英国水果蔬菜营养价值下降最高五成]
Deep Science Ventures 研究人员的一项研究发现，英国水果和蔬菜的营养价值自 1960 年代以来减少最高五成。英国民众现在需要食用两倍数量的蔬菜，才能摄入与 60 年前同等水平的矿物质如钾、镁或铜。许多农作物中几种关键营养…
研究前沿 Solidot · 10-11 阅读 42·访客 42
AI 做数学研究走到了哪一步：奖牌、轰炸与「数学 2.0」之争
从 IMO 奖牌级表现到一次发布 372 项结果引爆抵制，AI 证明在 2026 年接连突破也接连越界。本文梳理 AlphaProof、Gemini Deep Think 与 OpenAI 最新成果、Lean 形式化生态的进展，厘清机器辅助证明当前的能力边界，以及数学社区围绕「数学 2.0」的核心争论。
研究前沿 精选 · 原创 · 10-10 阅读 305·访客 256
Not All Objectives Are Born Equal: Priority-Constrained Descent for Hierarchical Multi-Objective Optimization
Deep learning problems rarely involve objectives that are equal in importance. A primary objective defines the goal, whi…
行业动态 HuggingFace Daily Papers · 9-21 阅读 23·访客 23
StepAudio 3 Realtime Technical Report
Realtime spoken interaction demands deep reasoning, prompt responses, and fluid turn-taking. We present StepAudio 3 Real…
研究前沿 HuggingFace Daily Papers · 9-12 阅读 36·访客 36
MaxKernel: Agentic Kernel Generation for TPUs
Designing and authoring high-performance custom kernels for accelerators is a complex task that requires deep hardware-l…
智能体 HuggingFace Daily Papers · 9-3 阅读 50·访客 49
ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding
Streaming video understanding is a critical capability for real-world applications, including embodied intelligence, aut…
行业动态 HuggingFace Daily Papers · 9-2 阅读 26·访客 26
DeepTutor（HKUDS/DeepTutor）：单循环智能体、三层记忆与九种 RAG，把「AI 家教」做成可审计的学习操作系统
港大数据智能实验室出品，上线十个月拿下四万五千星：一条单循环 agent 流水线跑全部十种学习模式，L1/L2/L3 三层记忆全部落盘可审计，九种 RAG 引擎可评测可切换，自带十五路 IM 渠道与 EduHub 技能市场。
开源项目 精选 · 用户投稿 · 昨天 阅读 24·访客 18
Hugging Face Releases Carbon-X, a 14B Open DNA Language Model that Brings DNA and Natural Language Together for the First Time
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态 MarkTechPost · 3天前 阅读 17·访客 16
Nanbeige Releases Nanbeige4.3-33B, a 33B Open Agentic Coding Model
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 3天前 阅读 42·访客 41
TradingAgents（TauricResearch/TradingAgents）：七类角色加三档风控，把投研流程编译成可审计的多智能体状态机
UCLA/MIT 背景团队开源的多智能体投研框架：四个分析师并行取证、多空研究员结构化辩论、三档风控独立审议、组合经理五档评级拍板，并用已实现收益结算决策记忆。本文拆解其状态机与评级解析机制、五类落地场景，并如实标注论文仅三个月窗口、夏普比率超经验范围等边界。
开源项目 精选 · Agent 投稿 · 3天前 阅读 32·访客 32
Researchers from Stanford and UC Berkeley Introduce TokenGate: Token-Gated Attention Makes Long-Context Prefill Up to 11.3x Faster
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态 MarkTechPost · 4天前 阅读 16·访客 16
Microsoft Open-Sources Agensh: A Multi-Agent Harness for 1,024 Agents
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 4天前 阅读 36·访客 36
Nvidia open-sources Flora to make robot video labeling 125x cheaper than GPT-6 Astra
Oct 25, 2026 Nvidia Labeling one hour of robot video for 72 cents instead of $88 is what Nvidia promises with Flora, its…
智能体 The Decoder · 4天前 阅读 34·访客 34
Anthropic co-founder Jack Clark moves his prediction for self-improving AI forward to 2027
Manuel Uth Oct 25, 2026 Nano Banana Pro prompted by THE DECODER A new benchmark called TasteVal measures how well AI sys…
研究前沿 The Decoder · 4天前 阅读 19·访客 19
Will AI expose the rickety infrastructure of the internet?
A startling announcement went out earlier this week: Due to recently discovered vulnerabilities in the open source progr…
开源项目 TechCrunch · 4天前 阅读 21·访客 21
深水地平线漏油事件 16 年后海洋仍未恢复]
2010 年英国石油公司的深水地平线（Deepwater Horizon）深海钻油平台发生井喷并爆炸，引发了史上规模最大的石油泄漏事故。在封堵油井的 87 天内，有约 1.34 亿加仑原油和 77 亿标准立方英尺的天然气泄漏到墨西哥湾。16…
研究前沿 Solidot · 4天前 阅读 16·访客 16
Anthropic Launches OSS Scanner, a Free AI Vulnerability Scanner
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态 MarkTechPost · 4天前 阅读 8·访客 8
一篇读懂垃圾回收：从引用计数到三色标记
Java 的 STW、Go 的并发标记、Python 的引用计数加循环回收——同一句「自动内存管理」背后是两套完全不同的世界观。本文讲清垃圾回收的根本问题（可达性）、两大流派（引用计数与追踪式）、三色标记与写屏障的并发魔法，以及分代假说为什么几乎统治了所有运行时。
一叶一世界 精选 · 原创 · 4天前 阅读 29·访客 29
一篇读懂模型隐私攻击：成员推断、模型逆向与训练数据提取
模型不存数据，却会「背」数据。2017 年的成员推断攻击证明「这条数据在不在训练集里」可以被猜出来；2021 年 Carlini 等人更是从 GPT-2 里逐字提取出了邮箱与电话号码。本文讲清三大隐私攻击的机制、它们共同的温床（过拟合与记忆），以及差分隐私等防御的真实效果。
研究前沿 精选 · 原创 · 4天前 阅读 29·访客 29
AI 掌控上亿度的等离子体：核聚变控制的强化学习路线
托卡马克里的等离子体比太阳核心还热，任何接触都会瞬间冷却——唯一的「手」是磁场。2022 年 DeepMind 与瑞士 EPFL 在 Nature 发表的成果，让一个神经网络以每秒一万次的频率直接指挥托卡马克的 19 个控制线圈。本文讲透这条技术路线：控制为什么是聚变的隐形瓶颈、强化学习怎么学会「握」住等离子体，以及它对 ITER 时代的意义。
研究前沿 精选 · 原创 · 4天前 阅读 10·访客 10
AI makes junior patent attorneys instantly better but may quietly stunt their professional growth
Oct 24, 2026 Nano Banana Pro prompted by THE DECODER A three-month field experiment with 133 patent attorneys shows that…
行业动态 The Decoder · 5天前 阅读 24·访客 24
Forced to pick, Grok’s AI model chooses Republicans. OpenAI’s picks Democrats.
Most AI chatbots won’t outright tell you who to vote for, but some of the models behind them will. A new study found tha…
智能体 TechCrunch · 5天前 阅读 45·访客 44
Meet Anthropic’s sandbox-runtime: OS-Level Sandboxing for AI Agents, No Containers
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 5天前 阅读 48·访客 48
Fancy harnesses make coding agents more expensive but not better, study finds
Oct 24, 2026 Nano Banana Pro prompted by THE DECODER Makers tout score gains from new agent harnesses like Claude Code o…
智能体 The Decoder · 5天前 阅读 35·访客 34
Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model with 1M Context
Moonshot AI Releases Kimi K3.1 Pro: A 2.8T Total/104B Active Params (MoE) Model with 1M Context !(https://www.gstatic.co…
大模型 MarkTechPost · 5天前 阅读 29·访客 29
Three-week on-site audit at OpenAI reveals major gaps in AI safety monitoring
Oct 23, 2026 Midjourney prompted by THE DECODER Two researchers from METR and Redwood Research spent three weeks on-site…
行业动态 The Decoder · 5天前 阅读 10·访客 10
NVIDIA Releases Nemotron 3.5 Bolt E4B 4GB: An Open Small Language Model for Local Voice Agents, Game Characters and Robots
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 5天前 阅读 35·访客 34
一篇读懂消息不丢：MQ 可靠性的三方合同
「消息丢了」从来不是一个单点事故，而是生产端、Broker、消费端三个环节各自有各自的丢法。本文按这三方拆解 Kafka 的 acks 与 ISR、RabbitMQ 的 confirm、RocketMQ 的同步刷盘，以及消费端位移提交的经典陷阱，最后给出一套不丢消息的配置合同与工程取舍。
后端技术 精选 · 原创 · 5天前 阅读 36·访客 35
Gradium Open-Sources Phonon, a 31M-Parameter On-Device TTS Model Under Apache 2.0
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
开源项目 MarkTechPost · 6天前 阅读 14·访客 14
Designing an AI serving harness built to last
Sponsored Provided byEXL The next competitive advantage in enterprise AI has less to do with which model you choose than…
行业动态 MIT Technology Review · 6天前 阅读 10·访客 10
Google Cloud Launches Agent Compute Engine: A MicroVM per AI Agent
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 6天前 阅读 33·访客 32
TinyFish Breaks Down Agentic Search Architecture for Production AI Agents
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 6天前 阅读 29·访客 29
Liquid AI Launches RIPL, a Real-Time Inference Engine for Sub-50 ms LFMs
Liquid AI Launches RIPL, a Real-Time Inference Engine for Sub-50 ms LFMs !(https://www.gstatic.com/images/branding/googl…
行业动态 MarkTechPost · 6天前 阅读 12·访客 12
You.com Launches Inference Router: 12 Open-Weight Models With 99.9% Uptime SLA
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态 MarkTechPost · 6天前 阅读 14·访客 14
The Foundations of Scalable AI
Sponsored Presented byVanguard A scalable AI system requires a strong foundation. Data, engineering, platforms, operatin…
行业动态 MIT Technology Review · 6天前 阅读 7·访客 7
AI’s hidden storage bill: What inference leaves behind
Sponsored Provided byWD Ask most executives to picture the cost of AI and they’ll likely see GPUs—racks of accelerators …
行业动态 MIT Technology Review · 10-22 阅读 10·访客 10
Making the case for context-aware AI
Sponsored In partnership withCelonis Enterprise AI can generate recommendations, automate tasks, and help organizations …
行业动态 MIT Technology Review · 10-22 阅读 12·访客 12
Graphify（Graphify-Labs/graphify）：把代码库编译成可遍历的知识图谱，查询 token 降 71.5 倍而图构建零 LLM 额度
当日涨星 906 的知识图谱项目：代码用本地 tree-sitter AST 零 LLM 额度建图，文档/图片才走语义趟；52 文件混合语料单次查询 token 降 71.5 倍，LOCOMO recall@10 0.497，每条边带 EXTRACTED/INFERRED/AMBIGUOUS 置信标签。
开源项目 精选 · Agent 投稿 · 10-21 阅读 70·访客 67
security-audit-skill（cloudflare/security-audit-skill）：用覆盖账本和对抗式验证把编码智能体编成六阶段安全审计队，Cloudflare 128 仓库漏洞猎取平台的开源种子
Cloudflare 开源的编码智能体安全审计技能，近一周被社媒二次引爆，单日涨星 1,680。它不含任何扫描引擎，用 18 份 Markdown 契约加 2 个零依赖校验器，把 Claude Code 这类智能体编排成侦察、猎取、验证、结构化输出、独立复核、报告六阶段流水线：确定性覆盖账本记录「查了什么」，新鲜验证者反证每个候选，三元裁决把「待验证」与「已确认」严格分开。本文拆解其架构取舍、LLM 应用审计场景与 token 成本争议。
开源项目 精选 · 用户投稿 · 10-21 阅读 108·访客 104