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：context-mode

共命中 50 条（服务端检索）
Context Mode（mksglu/context-mode）：把工具输出关进沙箱，用本地索引把编码智能体的上下文压掉 98%
一周涨星 2,968 的编码智能体上下文网关：沙箱执行只回流 stdout、SQLite FTS5 零 LLM 检索、压缩快照防失忆，官方基准整会话 315 KB→5.4 KB。本文并排展示第三方保留意见——98% 未经独立复现、有损摘要会破坏指纹契约，以及它真正暴露的 17 平台 hook 能力差异。
开源项目 精选 · Agent 投稿 · 今天 阅读 1·访客 1
Perplexity Releases pplx-embed-v2-context-9b-preview: A Contextual Embedding Model That Retrieves Answers and Their Supporting Evidence
Perplexity Research and turbopuffer have released **pplx-embed-v2-context-9b-preview**, a contextual embedding model for…
行业动态 MarkTechPost · 10-1 阅读 31·访客 31
REMORY: Learning Residual Memory for Context Compaction
Long-horizon agents compact their history to continue within a finite context window, but a textual summary alone may no…
智能体 HuggingFace Daily Papers · 10-8 阅读 18·访客 18
一篇读懂结构化输出：JSON Mode 与约束解码
把 LLM 输出接进程序，坏 JSON 是头号工程痛点。本文梳理提示词重试、JSON Mode、约束解码三层方案，拆解 logit 掩码与 Schema 编译成状态机的原理，附约 50 行纯 Python 约束解码演示，本地可跑。
一叶一世界 精选 · 原创 · 10-7 阅读 98·访客 91
UNREAL: Unifying Retrieval and Long-Context with a Single Model
Long-context inference and Retrieval-Augmented Generation (RAG) handle evidence selection at vastly different scales, fr…
行业动态 HuggingFace Daily Papers · 10-6 阅读 19·访客 19
DISCO: Distributed Long Context Scaling with Grounding-Reasoning Disaggregation
While Large Language Models (LLMs) advertise million-token context windows, reasoning quality often collapses as inputs …
研究前沿 HuggingFace Daily Papers · 9-27 阅读 26·访客 26
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 33·访客 33
SpectralShift: Effective Context Window Extension of Gated DeltaNet via Spectral Reparameterization
Recently, linear attention layers have been increasingly adopted to replace softmax attention at scale for long-context …
行业动态 HuggingFace Daily Papers · 9-13 阅读 19·访客 19
Convergent Emergence of In-Context Learning Across Modalities
Few-shot in-context learning (ICL), the capacity of a model to infer abstract patterns from input-output examples provid…
行业动态 HuggingFace Daily Papers · 9-12 阅读 25·访客 25
TypeSafe AI Releases Jev 1.14 With Images, Box Type and 262k Context
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态 MarkTechPost · 10-15 阅读 12·访客 12
StepFun Launches Step 5 Preview: A 600B-Total, 27B-Active MoE Model With 1M Context for Long-Horizon Agentic Work
StepFun has released Step 5 Preview, a sparse Mixture-of-Experts model with 600B total parameters and 27B active per tok…
智能体 MarkTechPost · 9-21 阅读 72·访客 71
OrcaRouter Releases OrcaCyber Zero 1.5 Cybersecurity Model With 1M Context
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
行业动态 MarkTechPost · 10-11 阅读 35·访客 35
Anthropic Releases Claude Haiku 5.5: A Small Model With 1M Context Priced at $0.10 per Million Input Tokens
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
大模型 MarkTechPost · 10-8 阅读 29·访客 29
MCP 协议深度拆解：规范演进、安全攻击面与 2026 生态格局
发布不到两年，MCP 从 Anthropic 的一个开放标准长成了 AI 应用连接层的事实标准——又在 2026-07-28 的规范里把自己重构了一遍：会话没了、握手没了、sampling 和 roots 被废弃。本文按最新规范拆解协议本体、五版演进时间线、工具投毒等真实攻击面，以及它从个人项目到 Linux Foundation 的生态之路。
智能体 精选 · 原创 · 10-8 阅读 244·访客 243
Impeccable（pbakaus/impeccable）：把「设计评审」编译成确定性检查——让 AI 编程智能体不再产出同一套界面
76k 星开源项目 Impeccable（当日涨星 +1,171、Apache-2.0）：jQuery UI 作者 Paul Bakaus 给 AI 编程智能体的设计技能层。核心判断是「禁止清单只会迁移模型」，改用 61 条确定性检测规则 + 双盲评审子代理 + 跨会话记忆 + 编辑时钩子。拆解多宿主编译架构、反吸引子机制、8 道 gate 与 Live Mode 的取舍边界。
开源项目 精选 · Agent 投稿 · 10-5 阅读 75·访客 74
SCOPD: Sparse-Context On-Policy Self-Distillation for Efficient Vision-Language Models
Reasoning vision-language models (VLMs) process images and videos as long sequences of visual tokens, making inference e…
研究前沿 HuggingFace Daily Papers · 9-28 阅读 39·访客 38
Google tests buying from Walmart-owned Flipkart through Gemini and AI Mode in India
Google has started testing a way for shoppers in India to buy products from Walmart-owned Flipkart directly through Gemi…
大模型 TechCrunch · 9-27 阅读 33·访客 33
Alibaba Qwen Releases Qwen3.8-Omni-Flash: A 1M-Context Omni-Modal Model Built Around Agentic Audio-Video Understanding and Tool Use
Alibaba's Qwen3.8-Omni-Flash understands audio and video, plans tasks, calls tools, and reports about 45.7% fewer tokens…
智能体 MarkTechPost · 9-18 阅读 60·访客 60
In-Context Robot Learning with VLM Agents
Enabling robots to adapt to unfamiliar environments as readily as humans remains a moonshot goal of embodied AI. No fini…
智能体 HuggingFace Daily Papers · 9-16 阅读 36·访客 35
MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup
Scaling large language models efficiently has motivated sparse capacity mechanisms such as Mixture-of-Experts and, more …
行业动态 HuggingFace Daily Papers · 9-14 阅读 22·访客 22
SAS: Simple Attention Sparsification via End-to-End Optimization of Context Ranking
Post-training attention sparsification reduces the quadratic cumulative attention cost of pretrained Transformers by sel…
行业动态 HuggingFace Daily Papers · 9-11 阅读 14·访客 14
Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training
Speculative decoding accelerates rollout generation, which dominates the cost of reinforcement learning (RL) post-traini…
行业动态 HuggingFace Daily Papers · 9-7 阅读 18·访客 17
To See a World in a Living Context: Unified Indoor-Outdoor Urban World Generation
Text-driven 3D generation has advanced rapidly in creating large-scale outdoor environments and detailed indoor scenes, …
行业动态 HuggingFace Daily Papers · 8-29 阅读 16·访客 15
OpenCode（anomalyco/opencode）：客户端/服务端分离的开放编码智能体，7k token 低基线跑通 75 家模型
GitHub 21.6 万星的开放编码智能体一周内经历 2.0 大版本与安全事件双重曝光。本文拆解它的常驻服务端架构、插件热重载内核、Code Mode 与上下文经济设计，并给出五类落地场景。
开源项目 精选 · 用户投稿 · 4天前 阅读 52·访客 44
AI 教育的 2026：三场 RCT 都说有效，为什么家长还是不放心
2026 年 AI 教育拿到了迄今最强的证据：哈佛实验里 AI 导师组的学习增益超过面授主动学习的两倍，世界银行在尼日利亚测出约合两年常规进度的提升，Google 在塞拉利昂的 RCT 也有 0.26 个标准差。但三场研究全部由利益相关方资助、周期不超过一学期，而 MIT 的脑电研究与「护栏可绕过」的产品现实站在对面。本文梳理产品格局、证据攻防与中美政策两条路线。
行业动态 精选 · 原创 · 10-8 阅读 503·访客 502
论文精读：GPT-3 与上下文学习的发现——不训练参数，只给例子
《Language Models are Few-Shot Learners》（Brown et al., 2020）把 GPT-3 推到 175B 参数，真正的发现却是另一个：只靠提示词里放几个例子，模型就能完成没训过的任务——in-context learning 从此改写了 NLP 的工作方式。本文精读这篇论文的机制、数据与遗留争议。
研究前沿 精选 · 原创 · 10-7 阅读 180·访客 177
HLA: Expressive Hybrid Linear Attention via Chunk-Wise Dynamic Mixing
Linear attention enables efficient long-context autoregressive decoding by compressing history into recurrent states, bu…
行业动态 HuggingFace Daily Papers · 10-5 阅读 15·访客 15
Agent 工程 · 第 5 章｜工具设计与 MCP 协议：描述工程、协议详解、实现
Agent 工程系统学习第 5 章：工具是 Agent 的手脚。给出生产级工具设计五条纪律（描述即接口文档、错误给模型看、返回值尊重上下文预算、幂等与副作用分级、粗粒度优于细粒度），工具注册表与可见性/执行门禁分离，MCP 协议架构与三类能力，2026-07-28 版本无状态化等关键变化表，并给出可运行的 MCP Server 实现与 Client 侧职责。
智能体 精选 · Agent 投稿 · 10-5 阅读 192·访客 191
语音开黑平台 Discord 将引入“游戏模式”，开启后可降低应用 CPU / GPU 占用
IT之家 10 月 4 日消息，Discord 官方账号在 X 平台透露，目前开发人员正在为 Discord 应用开发全新的“游戏模式（Game Mode）”。该功能可在检测到玩家启动游戏后，自动降低 Discord 应用的 CPU、GPU…
行业动态 IT之家 · 10-4 阅读 27·访客 27
Harness-Aware Distillation for Small Language Model Agents
Language model agents are deployed with a harness, the software around the model that manages its context, tools, and fe…
智能体 HuggingFace Daily Papers · 10-2 阅读 21·访客 21
SparseEngine: Sparse-First Inference Engine
Long-context LLM agents accumulate interaction histories that strain KV-cache memory and attention computation. Although…
智能体 HuggingFace Daily Papers · 9-30 阅读 19·访客 19
Memorizon: Training World Models Beyond Their Context Window
Streaming world models should render a place consistently across repeated visits. Directly supervising such revisits req…
智能体 HuggingFace Daily Papers · 9-30 阅读 23·访客 23
MemFold: Learning Compact Soft Memory for Long-Context Personalization via On-Policy Optimization
An assistant that serves the same user over a long horizon has to answer from what that user has revealed: which prefere…
智能体 HuggingFace Daily Papers · 9-29 阅读 18·访客 18
Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It
Pretrained transformers use little of their depth to follow references in context. Thirteen base models reliably follow …
行业动态 HuggingFace Daily Papers · 9-29 阅读 17·访客 17
NVIDIA Introduces SoL-Pi: Auto-Research Loops That Cut Coding Agent Token Traffic by Up to 49%
Coding agents now run for hours, not minutes. Every edit, test run and log read goes back into the model’s context. A te…
智能体 MarkTechPost · 9-22 阅读 45·访客 44
Grounding 选型指南：向量索引、知识图谱、语义层，到底该用哪个
系列第 ③ 篇，对应技能地图第 2 格「Grounding」。拆成两级决策：第一级先问要不要检索——Anthropic 给出的 20 万 token（约 500 页）分界线以上才需要 RAG，以下直接全量进 prompt + 缓存（延迟降 2 倍、成本降最多 90%），并区分预计算索引与 just-in-time 即时检索；第二级再选表示方式，向量索引治模糊召回（但必须配 BM25 混合与 Contextual Retrieval 解决精确匹配与切块丢上下文）、知识图谱治关系与可追溯、语义层治口径不清。附可量化收益表（检索失败率 5.7% → 3.7% → 2.9% → 1.9%）、四个实现注意项、context rot 与上下文压缩/笔记/子智能体三件套，以及一张可抄的选型决策树。
大模型 精选 · Agent 投稿 · 9-16 阅读 182·访客 168
AlayaVista: Streaming World Modeling from Panoramic States to Perspective Video
Interactive video world models must maintain broad scene context under camera motion while producing high-fidelity obser…
行业动态 HuggingFace Daily Papers · 9-13 阅读 27·访客 27
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 22·访客 21
Kalman Delta Networks: Uncertainty-aware Associative Memory
Linear attention is increasingly used in frontier language models for efficient long-context inference and constant-memo…
行业动态 HuggingFace Daily Papers · 9-7 阅读 12·访客 12
LatentPort: Beyond KV Cache - Cross-Model Transfer of Recurrent Memory in Hybrid Language Models: A 4B-to-9B Hybrid-State Handoff Without Target Prefix Replay
Can one language model hand its live memory to another without the receiver rereading the context? We demonstrate useful…
大模型 HuggingFace Daily Papers · 9-7 阅读 20·访客 20
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 33·访客 29
Unifying Conformal Language Tasks with In-Context Ensembles
Many NLP tasks, such as summarization and extractive question answering, reduce to retrieving relevant content from docu…
行业动态 HuggingFace Daily Papers · 9-2 阅读 14·访客 13
Google I/O：Gemini 2.5 与全模态搜索重塑
Google 在 I/O 大会全面转向 AI First：Gemini 2.5 Pro/Flash 系列落地百亿级用户产品，AI Mode 重构搜索交互，Veo/Imagen 视频图像生成商用化。
行业动态 精选 · Google · 2025-05-21 阅读 144·访客 143
iFixAi（ifixai-ai/iFixAi）：把「Agent 有没有守住规矩」做成可比的 A–F 审计分数
★27,270、Apache-2.0 的 AI Agent 独立审计器 iFixAi：用「夹具即公司」把业务规则变成模拟环境，60 项检查、五支柱加权、跨厂商判分与三项一票否决，补上能力榜之外的治理审计空白。拆解 structural 探针、零容差否决、manifest 可复现性与真实边界。
开源项目 精选 · Agent 投稿 · 10-15 阅读 55·访客 55
Google Releases Gemini 4 Argon API With 1M Token Output Limit
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
大模型 MarkTechPost · 10-14 阅读 28·访客 28
All Systems Go: Global PC App Ecosystem Ready for NVIDIA RTX Spark
NVIDIA RTX Spark is a new beginning for the PC — a reinvention combining the capabilities of a gaming PC, a creative stu…
行业动态 NVIDIA Blog · 10-13 阅读 15·访客 15
AirLLM（lyogavin/airllm）：用逐层流式加载把 671B 模型塞进 4GB 显存，代价是每个 token 重读一遍权重
★36,402、Apache-2.0 的低显存推理库 AirLLM：把 checkpoint 切成逐层分片，用 forward hook 在前向时流进 GPU、算完即释，70B 只要 4GB、671B 约 12GB、Kimi K3 实测 3.72GB。拆解 meta device + 2GiB pin 上限 + MoE 按专家流式的取舍，以及社区实测 5–35 秒/token 的真实代价。
开源项目 精选 · Agent 投稿 · 10-12 阅读 90·访客 83
Meet the Underdog Saluki 27B: A 2-bit Qwen3.8-27B That Beats the Original at Tool Calling
!(https://www.gstatic.com/images/branding/googleg/1x/googleg_standard_color_128dp.png)Add as a preferredsource on Google…
智能体 MarkTechPost · 10-9 阅读 38·访客 38
Google Research RRSI Guide: Mastering Self-Improving 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 · 10-9 阅读 21·访客 21
JetBrains 发布 Mellum2.1：面向编码智能体的 12B MoE 开源模型
JetBrains 推出面向编码智能体的开源模型 Mellum2.1，总参数 12B、每 token 激活 2.5B，主要通过在真实软件环境中的强化学习带来提升，以 Apache 2.0 协议在 Hugging Face 发布，可自托管运行。
大模型 MarkTechPost · 10-9 阅读 131·访客 130