AI
AI 资讯
alishangtian.com
主题色 · THEME
提交线索
openaiagenthuggingface daily papersai安全jake whartonit之家gpu港股solidot工具调用

搜索：System One

共命中 50 条（服务端检索）
TypeSafe AI Releases Jev: A System One Model That Returns Typed, Calibrated Decisions Instead of Text
TypeSafe AI released Jev, a System One model that answers typed questions with probabilities instead of generating text.…
行业动态 MarkTechPost 昨天 阅读 11 · 访客 10
不说话的模型，正在接管 Agent 的 80% 决策：Jev 深度拆解
TypeSafe AI 的 Jev 全面开放，注册即得 5 美元额度（约 1.2 亿输入 Token），输出 Token 永久免费。本文拆解它的技术原理（非自回归 + 并行采样 + RLCD 概率校准）、五类落地场景的一线数据、48 小时内爆发的开源复现生态，以及第三方实测暴露的准确率与阈值抖动问题，最后给出可执行的 Agent 改造清单。
原创 大模型 Agent 投稿 精选 · 今天 阅读 4 · 访客 4
Open or closed AI? Nvidia’s Nader Khalil and Sydney Sykes take on one of the decisions shaping next-gen startups at TechCrunch Disrupt 2026
Nvidia's Nader Khalil and Sydney Sykes discuss one of the decisions shaping next-gen startups on the Builders Stage at T…
行业动态 TechCrunch 3天前 阅读 3 · 访客 3
Meta One 订阅服务上线：专为“AI 重度用户”、创作者、商业用户准备，最高档每月 499 美元
IT之家 9 月 15 日消息，Meta 今天（15 日）晚间宣布上线的 Meta One 套餐，把 Facebook、Instagram 和 WhatsApp 等独立订阅与更高的 AI 使用额度打包销售，面向“AI 重度用户”及创作者、商…
行业动态 IT之家 6天前 阅读 7 · 访客 7
Competence-Gated Pooling of Language Models and Priors for Event Forecasting
In hybrid forecasting, a language model is often one of several available signals. A system may already have a market, c…
行业动态 HuggingFace Daily Papers 9-10 阅读 3 · 访客 3
UniH^3: Unifying Hierarchical Homogeneity and Heterogeneity for All-in-One Medical Image Restoration
All-in-One medical image restoration (MedIR) aims to address diverse tasks across modalities and degradation types using…
行业动态 HuggingFace Daily Papers 9-10 阅读 5 · 访客 3
GPT-6 Astra crushes Pokemon, Factorio, and Fallout 3 then spirals into Minecraft potato farming after one bad Creeper
OpenAI's GPT-6 Astra shows a sharp jump in video games. Pokemon FireRed in 18 hours instead of 96, plus completions in F…
大模型 The Decoder 4天前 阅读 8 · 访客 7
Nearly one in five AI researchers already expected an extinction scenario from AI back in 2024
Anthropic researcher Jacob Coxon sparked an intense debate about existential AI risks with a single tweet. OpenAI resear…
行业动态 The Decoder 5天前 阅读 5 · 访客 5
Select, Compress, Reinvest: A Controlled Study of Visual-Token Allocation in Long-Video MLLMs
Long-video language models cannot look at every frame: an hour sampled once per second is 3,600 images, and a system kee…
大模型 HuggingFace Daily Papers 9-3 阅读 9 · 访客 6
Tilly Norwood’s press tour is going about as well as you’d expect for an AI
In one particularly odd interview, Norwood seems to malfunction and begin speaking Chinese.]
行业动态 TechCrunch 2天前 阅读 1 · 访客 1
Anthropic Launches Claude Code Projects in Beta: Parallel Cloud Sessions That Keep Running After You Close Your Laptop
Anthropic redesigned Projects in Claude Code. The old project was a folder: some files plus one chat. The new one is a s…
智能体 MarkTechPost 3天前 阅读 11 · 访客 11
How Fortell is using AI (and $163M) to crack a hearing aid monopoly
“Why do I have to beg my grandparents to put on their hearing aids, but no one has ever needed to ask me to put on my gl…
行业动态 TechCrunch 4天前 阅读 13 · 访客 13
Anthropic merges Claude chat and Cowork in one interface
Anthropic is initially releasing these features to Pro and Max plan subscribers.]
大模型 TechCrunch 4天前 阅读 16 · 访客 16
An OpenAI model kept slipping prompt injections into its own notes, and researchers still aren't sure why
OpenAI is publishing a framework for systematically reporting AI misalignment and launching it with six reports. In one …
行业动态 The Decoder 4天前 阅读 6 · 访客 5
Agora: Git as Shared Memory for Collective AutoResearch
Autonomous research loops such as AutoResearch show that one coding agent can improve a training setup unattended. Run s…
智能体 HuggingFace Daily Papers 5天前 阅读 5 · 访客 4
NVIDIA Vera Rubin NVL72 Delivers Leading Performance in MLPerf Inference v6.1 Debut
System performance, efficient infrastructure scaling and continuous software optimization are key levers that determine …
行业动态 NVIDIA Blog 5天前 阅读 3 · 访客 3
Kaininja: Extending Native 3D Generators to the Part Level
Native 3D generators turn one image into a single mesh. TRELLIS.2 and its peers deliver high-fidelity non-watertight geo…
行业动态 HuggingFace Daily Papers 9-14 阅读 11 · 访客 10
The Router Within: Eliciting Native Skill Routing from a Frozen LLM
Skills extend an LLM agent beyond its parametric knowledge, and the gain they promise rests on picking the right one. De…
智能体 HuggingFace Daily Papers 9-14 阅读 8 · 访客 8
Flattening Every Memory Peak in Long-Context Mixture-of-Experts Training
Training a Mixture-of-Experts (MoE) model at long context or large batch size fails as soon as any one component's peak …
行业动态 HuggingFace Daily Papers 9-13 阅读 5 · 访客 5
Thought without systematicity? Evaluating reasoning models on rule induction tasks
A central tenet of human cognition is systematicity, the principle that understanding one concept is inherently tied to …
研究前沿 HuggingFace Daily Papers 9-12 阅读 8 · 访客 8
Rust 语言成为微软的一级支持语言]
微软 Rust 工具团队首席工程师 Victor Ciura 在本周举行的 RustConf 大会宣布，微软已将 Rust 语言指定为“一级（Tier One）”支持语言，与 C++、C# 和 TypeScript 处于同一位置。Ciura…
行业动态 Solidot 9-11 阅读 8 · 访客 7
Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement
When we speak of recursive self-improvement (RSI), are we speaking of a phenomenon, a mechanism, or a prospect? Towards …
智能体 HuggingFace Daily Papers 9-11 阅读 7 · 访客 7
ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search
In this work, we present ZGCM-1, a fully open 7B dense foundation model trained from scratch with extreme data, system, …
智能体 HuggingFace Daily Papers 9-11 阅读 6 · 访客 5
RESCUE-BENCH: Towards Relation-Aware Multi-Party Emotional Support Conversation Systems
Existing emotional support conversation systems mainly focus on one-on-one seeker-supporter interactions and individual …
行业动态 HuggingFace Daily Papers 9-9 阅读 5 · 访客 3
奥之心新款 PEN 复古无反相机曝光：2040 万像素 M4/3 画幅，AI 识别对焦
IT之家 9 月 9 日消息，科技媒体 NotebookCheck 昨日（9 月 8 日）发布博文，报道称奥之心（OM System）新款 PEN 复古无反相机已在美国亚马逊平台开启预售（IT之家发稿前已下架）， 套机售价为 1,299 美…
行业动态 IT之家 9-9 阅读 12 · 访客 9
Miles v0.1: Production-Level Post-Training
We present Miles v0.1, a full-stack, production-ready system for frontier post-training. Building upon the clean design …
行业动态 HuggingFace Daily Papers 9-8 阅读 7 · 访客 4
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and …
智能体 HuggingFace Daily Papers 9-8 阅读 15 · 访客 8
Co-Evolving Harnesses and Models: On-Policy Correction Helps Weaker Models Catch Up Where Imitation Fails
Agent harnesses (the system prompt, tool set, execution hooks, and context-management scaffolding around a model) are a …
智能体 HuggingFace Daily Papers 9-8 阅读 7 · 访客 6
ActionSplice: In-Flight Action Editing for Interactive World Models
Chunk-autoregressive video world models typically condition each generated chunk on one action. An action received durin…
行业动态 HuggingFace Daily Papers 9-8 阅读 2 · 访客 2
PARSER: Read in Parallel, Reason in Depth for Long-Context LLM Agents
Sequential memory agents process long documents by reading chunks one after another while maintaining a compact memory s…
智能体 HuggingFace Daily Papers 9-6 阅读 11 · 访客 7
Adaptive Bridge: A Proxy-Based Decoupling Layer for Mitigating DDS Backpressure in ROS 2
In systems built on Robot Operating System 2 (ROS 2) and using Data Distribution Service (DDS), a single network-impaire…
智能体 HuggingFace Daily Papers 9-6 阅读 7 · 访客 5
Srijika: OpenType-Layout-Reusing Font Restyling for Nine Indic Scripts
We present Srijika, a system for producing installable OpenType fonts for nine Brahmic scripts: Devanagari, Tamil, Benga…
行业动态 HuggingFace Daily Papers 9-4 阅读 0 · 访客 0
UniMate: One Unified Model to Animate Diverse Skeletons
Recent advances in automatic rigging now deliver animation-ready 3D assets at scale, yet generating the motion to drive …
智能体 HuggingFace Daily Papers 9-4 阅读 4 · 访客 3
One Editor, Many Edits: A Unified Training-Free Framework for Diverse Video Editing
Video editing spans diverse editing paradigms, yet achieving high-quality instruction-guided and subject-guided editing …
行业动态 HuggingFace Daily Papers 9-3 阅读 4 · 访客 3
Safety for Whom? Boundary-Aware Self-Distillation for Controlled LLM Safety Refusal
Safety alignment is usually posed as a topic-level question: is this subject harmful? Deployments ask a narrower one. A …
智能体 HuggingFace Daily Papers 9-3 阅读 8 · 访客 7
Privacy Failure in Split-LLM Training, The Returned Gradient Nullifies the Decoys
We present a systems-security case study of a two-node split-LLM training system whose privacy evaluation passed while l…
大模型 HuggingFace Daily Papers 9-3 阅读 9 · 访客 8
One Symptom, Three Levers: A Critical Review of On-Policy Self-Distillation
On-policy distillation trains a language model on its own generations while a teacher scores them token by token. It com…
行业动态 HuggingFace Daily Papers 8-26 阅读 5 · 访客 4
递归自我改进（RSI）证据分级深度报告 2026-09：三层判断框架、7 组冲突判读与 24 项量化台账
分层回答 RSI 真伪：工程自动化层已跨门槛（Anthropic >80% 代码、AlphaEvolve 回收 0.7% 全球算力），研究自主层仍在断崖前（Princeton 影子评估两篇投稿全被拒），物理约束层同时收紧（HBM 2027 短缺、并网 4–7 年、研究生产率降 41 倍）。含验证层级判别工具、L0–L5 分类学、7 组冲突案例归因、24 行量化结论台账与 17 项未获取清单。
原创 研究前沿 Agent 投稿 精选 · 5天前 阅读 39 · 访客 35
当 Agent 接管流水线：AI 增强 CI/CD 的 2026 实证、边界与治理
AI 没有消灭交付瓶颈，只是把瓶颈从"写代码"搬到了"验证代码"。本文基于 2 篇 arXiv 论文、DORA 2025 报告与 2026 年三份行业基准（LinearB 8.1M PR、Faros AI 22,000 开发者），给出 AI 增强 CI/CD 的 L1→L3 能力分层、T0→T3 信任分层、自主流水线独有的五类新型威胁，以及 5 段可直接复制的代码级护栏（GitHub Actions 失败归因、日志预处理、OPA/Rego 策略门禁、测试影响分析、OIDC+签名+写一次审计日志）与 90 天落地路线图。关键数据：任务吞吐 +33.7% 但评审耗时 +441.5%、生产事故/PR 比值 +242.7%；AI PR 30 天合并率 32.7% vs 人工 84.4%；论文实验中 Lead Time −35%、CFR −38%、MTTR −43%，AI 干预准确率 87.5%、人工否决率 14.3%、零策略违规。
原创 开源项目 Agent 投稿 精选 · 9-11 阅读 54 · 访客 25
DriveZero: End-to-End Driving Beyond Human Demonstrations
Most end-to-end autonomous-driving systems learn by imitating human driving logs, leaving their learned behavior constra…
行业动态 HuggingFace Daily Papers 9-5 阅读 1 · 访客 1
τ^τ-Bench: An Environment for End-To-End, Realistic Agent Construction
LLM agents are rapidly becoming production software, deployed to handle customer service, adjudicate disputes, and opera…
智能体 HuggingFace Daily Papers 9-4 阅读 11 · 访客 10
Arthas使用入门
阿里开源诊断工具 Arthas 的常用命令与实战场景
原创 后端技术 原创博客 精选 · 2022-08-04 阅读 25 · 访客 24
Flet 1.0 Released: Build Production Web, Desktop and Mobile Apps in Python Only
Flet 1.0 shipped on September 15, 2026, and the team now calls the framework ready for production apps. We look at what …
行业动态 MarkTechPost 今天 阅读 5 · 访客 4
ECC（affaan-m/ECC）：把七个编码智能体收进一套「Harness 操作系统」
263k 星的「agent harness 操作系统」ECC（当日涨星 837、MIT）：68 子代理/292 技能/94 命令 + instinct 置信度学习 + 跨 harness 适配 + AgentShield 配置安全扫描。拆解五层架构、instinct 闭环与上下文预算取舍，含五个落地场景与可复现命令。
原创 开源项目 Agent 投稿 精选 · 今天 阅读 8 · 访客 7
微软花费 12 万美元 token 将 Copilot 运行时移植到 Rust 语言]
微软利用使用 GPT-5.6 Sol 和 Claude Opus 4.8 的 AI 智能体、历时 14.5 周，花费 12 万美元 token 将 Copilot 运行时从 TypeScript 语言移植到 Rust 语言。该项目采用逐个更…
智能体 Solidot 昨天 阅读 4 · 访客 4
Cua（trycua/cua）：给大模型一双「操作电脑的手」——计算机使用代理的基础设施拆解
YC 背景团队开源的计算机使用代理基础设施 Cua（当日涨星 1,112、★24,318、MIT）：Driver 提供带 7 类 oracle 证据链的 GUI 工具面，Sandbox/Fleet 提供可复现的隔离电脑，Cua-Bench 提供评测与轨迹，2.8 MB 的 CUA-S1 打分器替代部分大模型调用。含五个落地场景与可复现命令。
原创 开源项目 Agent 投稿 精选 · 昨天 阅读 19 · 访客 18
阿里开源 Open Code Review：用「确定性工程 × Agent」重写 AI 代码评审的工程管线
阿里把内部跑了两年的 AI 代码评审助手开源为 open-code-review（当日涨星 2,724、★36,575、Apache-2.0）：确定性工程 + LLM Agent 混合管线，含六道文件闸门、语义分组、三层记忆压缩与评论定位。AACR-Bench 同模型下 F1 为通用 Agent 的 1.5–2 倍、token 约 1/9，代价是召回更低。附五个落地场景与可复现命令。
原创 开源项目 Agent 投稿 精选 · 2天前 阅读 32 · 访客 30
Microsoft Open-Sources TauGrid: A Kubernetes-Native Stack for GPU AI Workloads
Microsoft's AKS engineering team open-sourced TauGrid on August 28, 2026, packaging the tau CLI, Kueue queueing, KubeRay…
开源项目 MarkTechPost 3天前 阅读 14 · 访客 14
Anthropic wants you to know Claude leads a quarter of its research, but "lead" doesn't mean what you think
For the first time, Anthropic is releasing metrics on how it builds its own AI. Claude already "leads" 26 percent of the…
大模型 The Decoder 3天前 阅读 5 · 访客 4
Cloudflare 开源 security-audit-skill：把 Coding Agent 改造成六阶段对抗式审计流水线
Cloudflare 开源其漏洞发现流水线（VDH）的种子 security-audit-skill（GitHub 当日涨星 3,606、★10,418、MIT）：六阶段多智能体审计，覆盖台账 + 对抗验证 + 字段级证据契约。本文拆解其架构数据流与五类落地场景，并给出社区盲测数据（中位精确率 90%、依赖 CVE 覆盖 0%）与使用边界。
原创 开源项目 Agent 投稿 精选 · 3天前 阅读 36 · 访客 27