独立 AI 资讯库
从自动抓取数据库实时读取，适合快速浏览、检索和后续扩展成完整资讯站。
arXiv:2608.07527v1 Announce Type: new Abstract: Long-document understanding requires models to find and combine evidence across many pages, layouts, tables, figures, and charts. Existing retrieval-augmented systems usually select evidence from a static index before generation, while recent agentic
arXiv:2608.07525v1 Announce Type: new Abstract: Hallucination remains a persistent challenge for Multimodal Large Language Models (MLLMs), severely limiting their reliability in high-stakes applications. Existing evaluations, predominantly based on static benchmarks, suffer from narrow taxonomical
arXiv:2608.07775v1 Announce Type: new Abstract: Mobile agents have achieved promising results on clean online benchmarks such as AndroidWorld, yet their performance often degrades sharply in real-world deployment due to environmental variations and imperfect interface conditions. In this work, we i
arXiv:2608.07762v1 Announce Type: new Abstract: LLM benchmarks can build an organization's reputation and attract customers, but only when results are transparent and verifiable. Unverified claims that DeepSeek R1 outperformed OpenAI's o1 contributed to market panic on January 27, 2025, when Nvidia
arXiv:2608.07743v1 Announce Type: new Abstract: Identifying a meaningful quantum speedup requires more than matching a classical problem to a familiar quantum primitive: the claim must preserve the task, respect access and output models, expose required promises, and remain within a defensible comp
arXiv:2608.07700v1 Announce Type: new Abstract: Translating a natural-language question into a SPARQL query that can be executed against a large knowledge graph requires resolving lexical ambiguity, grounding surface terms in the target ontology, and producing graph patterns that are both syntactic
arXiv:2608.07688v1 Announce Type: new Abstract: IT audits require auditors to judge whether heterogeneous organizational evidence satisfies semantic security and compliance controls. This judgment is difficult to automate because relevant evidence is distributed across policies, records, spreadshee
arXiv:2608.07651v1 Announce Type: new Abstract: Large language models (LLMs) show promise in medical image interpretation but suffer from hallucination, limited accuracy, and run-to-run inconsistency. We developed and validated an agentic AI framework integrating LLMs with specialized deep learning
arXiv:2608.07645v1 Announce Type: new Abstract: Self-improving coding agents that iteratively rewrite their own source code have demonstrated impressive performance on coding tasks. However, existing solutions generally derive self-modification from a single failure trajectory at a time, overlookin
arXiv:2608.07642v1 Announce Type: new Abstract: Aligning large language models (LLMs) with human values remains a major challenge, especially for trustworthy AI. While existing approaches such as RLHF, CAI, and their variants have achieved promising results, they often rely on a single-agent framew
arXiv:2608.07637v1 Announce Type: new Abstract: Long-running molecular simulation campaigns require repeated continuation from saved states, provenance-aware progression, adaptive assessment, and occasional interpretation of workflow conditions that cannot be resolved safely by fixed rules. Here, w
arXiv:2608.07627v1 Announce Type: new Abstract: Hospitals are racing to embed AI, while coping with the surge in adaptation of the technology in other industries, into the triage management, documentation, scheduling, and revenue-cycle workflows, yet most deployments remain as fragmented pilots tha
arXiv:2608.07622v1 Announce Type: new Abstract: Long-term memory enables AI agents to maintain continuity across sessions, personalize behavior, and evolve through accumulated experience. Yet memory evolution is not simply a process of storing more information: new experiences may reinforce, revise
arXiv:2608.07621v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have recently achieved impressive performance for end-to-end autonomous driving, yet existing approaches are primarily designed for an individual single autonomous driving agent with limited support for cooperative
arXiv:2608.07617v1 Announce Type: new Abstract: Scientific and technical writing depends on markup sources that must compile: LaTeX, Typst, and Markdown pipelines fail on missing delimiters, mismatched environments, broken imports, or package conflicts. Existing document-repair evaluations inject f
arXiv:2608.07540v1 Announce Type: new Abstract: AI systems increasingly operate between flexible input representations and formal objects used by downstream tools. A key challenge is recognizing when an unfamiliar formulation denotes a known formal object. We study this challenge through theorem re
arXiv:2608.07538v1 Announce Type: new Abstract: As LLM agents move from decision support to autonomous procurement, firms need to know whether delegated negotiators create value, divide it predictably, and avoid money-losing contracts. We study this in a canonical supply chain bargaining problem: a
arXiv:2608.07533v1 Announce Type: new Abstract: An embodied agent is an intelligent entity that interacts with its environment through a physical body. Currently, the evaluation of embodied agents primarily relies on two paradigms: (1) manually annotated Visual Question Answering (VQA) pairs and (2
arXiv:2608.07532v1 Announce Type: new Abstract: Modern agentic AI systems combine multiple large language model agents with heterogeneous skills, yet most architectures either fix communication in advance or allow full broadcast. Both can be inefficient because token cost, latency, redundancy, and
arXiv:2608.07530v1 Announce Type: new Abstract: SHACL is a core technology for validating the conformance of RDF knowledge graphs (KGs). Yet, authoring SHACL shapes requires technical expertise that most domain experts lack. Translating natural language requirements into SHACL (NL2SHACL) would lowe
jeeflow-go 面向需要在 Go 服务中接入审批流等工作流能力的开发者，主打轻量化部署与快速上手。开发者可通过一行 `go get` 安装，并使用内存仓储在约 5 分钟内跑通一条审批流程。其引擎核心不依赖第三方组件，生产环境则可通过统一的 45 个 action 对接 MySQL 等基础设施。原文声明其中的命令、代码和输出均来自真实机器实录，但当前提供的正文仅显示等待提示，未包含具体技术内容。
点击查看原文>
Pathway 推出的 Baby Dragon Hatchling（BDH）是一种后 Transformer 架构，可在潜在空间中进行循环推理，而非生成冗长的思维链 token。基于 BDH 的视觉推理系统 BDH-CQ 在 ARC-AGI-1 基准上取得 29.2% 的 pass@2，单任务成本为 0.0007 美元，并在截至 2026 年 8 月的成本—准确率帕累托前沿上实现突破。Pathway 使用 Amazon SageMaker HyperPod、EFA、EC2 p5en.48xlarge 实例及可观测性工具扩展其分布式训练工作负载。该案例表明，面向长程推理、低成本推理和持续状态管理的新型模型架构，可能为开发者提供 Transformer 之外的工程路径。
Meta 推出个人 AI Agent Muse，定位为一款安全、私密且可主动协助用户完成目标、提出建议的产品。Muse 运行在专属的 Muse Secure VM 上，可执行发送邮件、预订旅行、填写表单、协商议价和购物结账等任务，并由 Meta 的 Muse Spark 模型驱动。该产品通过 Sentinel Agent 审批联网操作、敏感操作确认、凭据隔离和完整审计记录等机制强化安全与隐私。Muse 已在美国面向 iOS、Android 和 muse.ai 推出，多数基础功能免费，并计划支持 AI 眼镜、Shop Pay、1Password 及端到端加密的 Confidential VM。
Meta推出面向美国消费者的个人AI Agent Muse，可连接电子邮件、日历、支付、健康、智能家居、购物等服务，代为完成发邮件、订旅行、填写表单和购买商品等任务。Muse基于Meta的Muse Spark模型运行，支持内置连接器、公共API接入及浏览器访问，并提供免费版以及每月20美元和100美元的付费套餐。Meta称Muse运行于独立的安全虚拟机中，无法查看用户密码或支付方式，也不会将对话和数据共享给其广告系统。由于Meta长期存在隐私争议、监管处罚和数据安全事故，Muse能否获得用户授权并建立足够信任，将成为消费级Agent落地的重要考验。
Amazon SageMaker Feature Store 新增 UpdateRecord API，允许开发者在一次调用中仅更新一个或多个特征值，无需读取或重写整条记录。该能力适用于基于 Amazon DynamoDB 的 Standard 在线存储层和基于 Amazon ElastiCache 的 In-Memory 在线存储层，其中 Standard 层需使用新的 Standard_V2 存储格式。新 API 通过原子合并减少读写延迟、RCU 成本及多流水线并发更新造成的丢失更新风险，同时支持基于 EventTime 的时序校验。AWS 还提供 Standard 到 Standard_V2 的批量迁移和原地切换方案，并新增 IAM 条件键以限制主体可更新的特征字段。
个人AI的实现，到底需要一套怎样的计算底座？当手机开始理解用户意图、调用工具并连续完成任务，计算平台要支撑的已不只是一次模型推理，而是完整的个人AI体验。在今天的Arm Everywhere China 2026大会上，Arm发布了最新的CSS for Mobile 2。可以观察到，AI在全新的手机计算子系统中扮演了两个不同角色，CPU通过增强AI计算能力，帮助更多应用用上AI；Mali GPU首次原生集成神经网络加速器，用AI推动游戏性能和体验跨越式提升。Arm的CPU正在推动端侧AI进一步普及。“目前几乎所有旗舰智能手机，无论是安卓还是iOS手机，均已采用SME2技术。”Arm边缘AI智能
过去，开源的定义很清晰，交出源代码，让所有人查看、修改、分发。如今，越来越多的开发者把精力从“写代码”转向“怎么让 AI 写出好代码”。他们调试提示词，打磨工作流，把与 AI 协作的一整套方法整理成文档，公开分享。分享的东西变了，开源的内核没变，把创造的过程分享出来，让后来的人少走弯路。 这正是 2026 北京...
点击查看原文>
MIT 一名研究人员展示了如何使用 GPT-5.6 Sol 与 Codex 自主运行量子计算实验。该系统可协助执行实验流程、分析实验结果，并开展量子比特校准。该案例表明，AI Agent 有望帮助研究人员自动化复杂的量子计算实验工作流，提高开发与科研效率。