Skip to main content

XMT

短闻

信流 · 上滑连读 · 来源可核

今日 稍后 搜索 RSS
1 / 24
Aggregate Hypebeast 时尚 45″

A24 Launches Pop-Up Store at Isetan Shinjuku Featuring 'Backrooms' and 'The Drama' Merchandise

SummaryA24 has opened a limited-time pop-up store at Isetan Shinjuku in Tokyo running through September 3 2026The event features collaborative silkscreen T-shirts with GEEKS RULE inspired by 'Backrooms' alongside logo apparel for 'The Drama'Additional items available include film soundtrack vinyl records and official studio merchandiseA24 and GEEKS RULE have officially launched an exclusive pop-up store at Isetan Shinjuku celebrating the releases of its films Backrooms and The Drama.…

  • Located on the second floor of the main building inside TOKYO Closet, …
  • Leading the apparel selection is a trio of graphic T-shirts created in…
  • Utilizing silkscreen printing techniques engineered to replicate the l…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate Hypebeast 时尚 45″

The New Balance 1890A "Cowprint" Wraps Its Loafer Silhouette in Spotted Fur for September

Name: New Balance 1890A “Cowprint”Colorway: Beige/BlackSKU: U18906M2MSRP: $190 USDRelease Date: September 16Where to Buy: New BalanceThe New Balance 1890A "Cowprint" takes the loafer-shaped model into fall with its heaviest texture yet.…

  • Built on an off-white suede base, the pair adds black-and-white spotte…
  • The story here is almost entirely about touch.
  • Where many colorways lean on contrast branding or bold palettes, the "…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate EE Times 芯片半导体 17″

TSMC’s HBM-Packaging Yield Issues Help Intel, Analysts Say

TSMC’s difficulty stitching together AI accelerators with high-bandwidth memory (HBM) on a single chip gives Intel a chance to win foundry customers, according to Counterpoint Research co-founder Neil

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate 雷锋网 人工智能 45″

越疆2026中报:营收翻倍、具身收入激增20倍,Physical AI进入商业化兑现期

当机器人行业的大量叙事还停留在融资轮次与概念发布时,越疆科技已经交出了一份由产线验收单写成的中期答卷。8月24日,越疆科技发布2026年上半年中期业绩。报告期内,公司实现营业收入约人民币3.…

  • 2亿元,同比增长106.
  • 6%,经营规模实现翻倍增长;毛利约人民币1.
  • 5亿元,同比增长约100%,毛利率稳定维持在47.

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Physics-Knowledge-Guided Hybrid Neural Learning for Arctic Sea Ice Concentration Evolution and Short-Range Prediction

arXiv:2608.21767v1 Announce Type: new Abstract: Accurate modeling of sea ice concentration (SIC) evolution is essential for polar climate assessment and short?…

  • range sea ice prediction.
  • Numerical and data-driven approaches constitute major foundations for …
  • tation, whereas the latter rarely encode physical dependencies explici…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

What Does CLIP Learn for Regional Geolocalization? Probing Visual Cues and Scene Configuration After Adaptation

arXiv:2608.…

  • 21761v1 Announce Type: new Abstract: Large collections of street-view …
  • In particular, fine-grained regional geolocalization is challenging be…
  • We study regional geolocalization within a metropolitan area and ask w…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

ECHO: A Cognitively Inspired, Auditable Memory Plane for Long-Horizon Agents

arXiv:2608.21755v1 Announce Type: new Abstract: Long-horizon agents need memory that identifies relevant experience, resolves revisions, and exposes checkable provenance.…

  • We present ECHO (Embodied Context and History Orchestration), an audit…
  • This is functional inspiration, not neural equivalence; the empirical …
  • Development runs reach 96.

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Ask or Answer: A Decision Framework for Multi-Turn Health Misinformation Intervention

arXiv:2608.…

  • 21721v1 Announce Type: new Abstract: Correcting health misinformation …
  • Yet existing methods either respond immediately or probe indiscriminat…
  • We propose Reward-Optimized Probe-and-Respond (RO-PnR), a framework th…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

ATHENA: Knowledge-guided agentic neural architecture search for AutoFormer-based electronic health record modeling

arXiv:2608.…

  • 21712v1 Announce Type: new Abstract: Transformer-based models are wide…
  • Neural architecture search (NAS) automates architecture design, but co…
  • Recent large language model (LLM)-guided NAS methods reduce manual sea…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate 数英网 时尚 45″

文章频道 - 德芙七夕文案,翻车了

在社交平台上掀起了一场关于品牌价值观的讨论。首发:GrayDesign近日,有网友发帖吐槽德芙官方旗舰店在七夕期间发布的一则短视频文案令人不适,并在社交平台上掀起了一场关于品牌价值观的讨论。…

  • 通过上图可以看到,文案写着:“跟男朋友要了很久的德芙七彩礼盒也没给我买,今天趁男朋友不在跟暧昧对象分享了抖音,驾车2小时送了过来,很甜。
  • ”视频标题还配上了“我看谁还没有收到男朋友送的德芙七彩礼盒”的话术。
  • 这段将“男友不送礼物、暧昧对象驱车2小时送礼”包装成“很甜”的剧情,将“暧昧越界”包装成浪漫桥段的文案,存在严重的价值观导向问题。

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate 数英网 时尚 45″

项目频道 - AI短片《反正也没时间活》,戳中了多少人的内心

一天24小时,除去工作、通勤和睡眠这些无法自由支配的,你还能有多少真正属于自己的时间?在抖音ai创作大赛中,一支片子《反正也没时间活》帮我们计算了,一生中自己可用的时间能有多少天。…

  • 反正也没时间活故事采用架空结构,虚构用金钱交易人生剩余时间的功能。
  • 男主背负房租、贷款和信用卡的压力,迫不得已向外星人出售自己的剩余时间。
  • 虽然他目前只有29岁,但往后一生中他能自由支配的时间仅有21天7小时。

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate Hypebeast 时尚 45″

OMEGA Brings the Speedmaster 38mm Line Closer to Classic Moonwatch DNA

SummaryOMEGA expands the Speedmaster 38 mm collection with seven new references inspired by the Moonwatch.…

  • Updates include rounded subdials, Dot‑Over‑Ninety bezels, cleaner dial…
  • Models span steel, bi‑colour, and diamond‑set editions, powered by Cal…
  • OMEGA has expanded its Speedmaster 38 mm collection with seven new ref…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate Solidot 综合科技 21″

15 年前库克接替乔布斯担任苹果 CEO

15 年前的 2011 年 8 月 24 日,蒂姆·库克(Timothy Donald Cook)接替史蒂夫·乔布斯担任苹果 CEO,下周的 9 月 1 日,库克将卸任 CEO 一职,John Ternus 将接替他。…

  • Ternus 于 2001 年加入苹果产品设计团队。
  • 他于 2013 年晋升为硬件工程副总裁,2021 年升任高级副总裁,期间负责监督了 AirPods、Mac、iPad 和 iPhone 的开发。
  • 库克将继续留在苹果公司,担任执行董事长,负责与政策制定者沟通,帮助苹果应对监管压力。

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate InfoQ中文 企业软件 6″

Grafana 正式发布 gcx 和 MCP 服务器,助力基于遥测的智能代理开发

点击查看原文>

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

From Association to Causation: Improving Retrieval Precision of Retrieval-Augmented Generation via Causal Relations and an Attention Mechanism

arXiv:2608.…

  • 21702v1 Announce Type: new Abstract: Retrieval-Augmented Generation (R…
  • We trace it to a conceptual gap: similarity captures only associationa…
  • We model the terminal retrieval stage with a causal graph grounded in …

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Context as an Environment: Programmatic Context Management for Long-Horizon Agents

arXiv:2608.21690v1 Announce Type: new Abstract: LLM agents increasingly take on long-running tasks whose history grows far beyond a single model context window.…

  • Existing approaches compress earlier interactions or extract selected …
  • We present Scroll, a context manager that treats each agent session as…
  • The environment is backed by an append-only Event Log and a sandboxed,…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

From Mastery Profile to Simulated Response: Stochastic Student Knowledge Graphs (SSKG) for Faithful LLM Student Simulation

arXiv:2608.21668v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used to simulate students at different mastery levels.…

  • These simulations can generate synthetic training data and stress-test…
  • However, common prompt-based approaches leave the answer decision to t…
  • As a result, these approaches may have difficulty distinguishing stude…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Measuring Activation Control in Large Language Models

arXiv:2608.…

  • 21664v1 Announce Type: new Abstract: Safe deployment of increasingly c…
  • However, if models can also control their own activations, deception c…
  • With this in mind, we introduce the Activation Controllability Benchma…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

SAEM: Stage-Aware Expert Management for Memory-Efficient MoE Inference in Chain-of-Thought Reasoning

arXiv:2608.…

  • 21614v1 Announce Type: new Abstract: Chain-of-thought (CoT) prompting …
  • Mixture-of-Experts (MoE) models scale capacity through sparse expert a…
  • Existing runtimes treat all tokens uniformly, overlooking a key struct…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate GQ 时尚 6″

Omega Just Shrank the Speedmaster to the Ideal 2026 Size

The iconic Moonwatch finally comes in more manageable proportions.

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Semantic Compression Trees: Multi-Resolution Knowledge Retrieval via Hierarchical Semantic Residuals

arXiv:2608.…

  • 21610v1 Announce Type: new Abstract: Retrieval-augmented generation re…
  • We introduce Semantic Compression Trees (SCT), a hierarchical index in…
  • We evaluate on QASPER (50 papers, 173 questions) under two protocols d…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Generate in the Chart, Not on the Boundary: Function-Symbol Grounding for Hard Constraints in LTN-GANs

arXiv:2608.…

  • 21605v1 Announce Type: new Abstract: Logic Tensor Network-Enhanced Gen…
  • Previous LTN-GAN work grounded every constraint this way, at the predi…
  • A predicate, however, only scores a sample, so it cannot embed hard st…

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

K-Bench: measuring model performance on real scientific agent requests

arXiv:2608.…

  • 21601v1 Announce Type: new Abstract: Benchmarks for scientific artific…
  • Real scientific requests arrive differently.
  • They are underspecified, they carry attachments, and lack ground truth.

RSS 官方收录 · 可信分层展示

详情 原文 分享图
Aggregate arXiv cs.AI 人工智能 45″

Data-Driven Dynamic Algorithm Dispatch with Large Language Models

arXiv:2608.…

  • 21584v1 Announce Type: new Abstract: We introduce a large language mod…
  • By combining prompt engineering with LLaMA 3 and a curated performance…
  • A case study on LU factorization demonstrates the model's ability to r…

RSS 官方收录 · 可信分层展示

详情 原文 分享图