微信内可能无法直接打开本站。请点右上角 ··· → 在浏览器打开,或复制链接后用系统浏览器访问。
XMT
信流 · 上滑连读 · 来源可核
Elie Saab Makes It Real With an H&M Collaboration That Brings His Red-Carpet Style Within Reach
Come October 22, the Elie Saab H&M collaboration at last puts Mr. Saab’s particular brand of glamour—spotted of late on Zendaya, Olivia Dean, and Eva Longoria—within reach for the rest of us.
RSS 官方收录 · 可信分层展示
跨境生意越做越“大”:Shopee卖家如何打开东南亚运动户外新机会?
2026年的大促旺季再次临近,过去几年,从居家健身到骑行、露营,运动正在更深地进入当地消费者的日常生活,随之变化的还有购物车里的商品:运动服饰、背包、水杯之外,跑步机、自行车、露营桌椅、帐篷等体积更大、客单价更高的商品,也开始承接越来越丰富的运动户外需求。…
- 这也让一些跨境卖家的选品思路发生变化。
- 过去做跨境,商品“小而轻”往往意味着更低的物流门槛和更灵活的经营方式;如今,随着东南亚消费需求不断细分、本地电商生态和履约能力逐渐成熟,一些…
- Andy(化名)正是提前押注这一变化的Shopee卖家之一。
RSS 官方收录 · 可信分层展示
READY or Not: Reliable Enterprise Agent Deployment
arXiv:2609.02095v1 Announce Type: new Abstract: An AI agent can perform well on benchmarks and still be unsuitable for deployment.…
- Existing AI-agent benchmarks measure whether an agent can complete rea…
- We introduce Reliable Enterprise Agent Deployment (READY), a framework…
- READY preserves each workflow's own definition of successful execution…
RSS 官方收录 · 可信分层展示
MASkills: Continual Skills Optimization for Multi-Agent LLM Systems
arXiv:2609.…
- 02094v1 Announce Type: new Abstract: LLM-based multi-agent systems hav…
- Existing self-reflection methods build experience memories, but memori…
- We introduce MASkills, a continual learning framework that optimizes m…
RSS 官方收录 · 可信分层展示
Beyond Outcome Gaps: Process-Aware Fairness Diagnosis for LLM-based Multi-Agent Decision Systems
arXiv:2609.…
- 02092v1 Announce Type: new Abstract: LLM-based multi-agent systems (MA…
- We present SCOPED-Hiring, a process-aware fairness diagnosis pipeline …
- SCOPED-Hiring constructs controlled resume variants, runs role-based h…
RSS 官方收录 · 可信分层展示
ToolGate: An Executable Acceptance Pipeline for Tool-Dependent Scientific Benchmark Construction
arXiv:2609.…
- 02067v1 Announce Type: new Abstract: Scientific benchmarks are commonl…
- These routes can produce strong evaluations, but they require substant…
- Language models can reduce this repeated work by proposing candidates …
RSS 官方收录 · 可信分层展示
CHIME: Credit-Aware Hierarchical Memory Evolution for Long-Horizon Agentic Planning
arXiv:2609.02074v1 Announce Type: new Abstract: Planning is a central capability that enables agents to decompose complex long-horizon tasks into manageable steps.…
- Test-time search and training-based methods improve planning but incur…
- Self-evolving memory instead accumulates reusable experience from agen…
- However, existing self-evolving memory methods share an inherent credi…
RSS 官方收录 · 可信分层展示
当 AI 开始理解「人不是标签」:阿里妈妈如何重构广告定向
作者|Cynthia 编辑|郑玄 2026 年夏天,林氏家居想卖一张布艺床。然后,阿里妈妈万相点睛给它找到了一群电竞玩家。并建议他们投放那些搜索过「电竞房布置」「游戏房改造」「RGB 灯带」「二次元痛房」的用户。…
- 对于林氏家居这样一家诞生于互联网、长期依赖数据经营的头部品牌来说,这是个让人匪夷所思的抉择。
- 过去,林氏家居品牌已经积累了超过 5500 万全平台粉丝、上万款在售 SKU 以及一千余家线下门店,是行业里公认的爆款制造机。
- 而在传统的匹配逻辑下,一张床最自然的消费者应该来自装修、新婚、换房,以及近期购买过家电家居的人。
RSS 官方收录 · 可信分层展示
MineTRACE: An Evidence-Grounded Interactive Reasoning System for Mineral Prospectivity
arXiv:2609.…
- 02060v1 Announce Type: new Abstract: Mineral exploration requires inte…
- We present MineTRACE, a web-based system for evidence-grounded explora…
- Users can explore prospectivity maps, query locations or regions, insp…
RSS 官方收录 · 可信分层展示
DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents
arXiv:2609.…
- 02059v1 Announce Type: new Abstract: Multimodal Large Language Models …
- However, existing benchmarks typically evaluate these domains in isola…
- We introduce DocHop, a benchmark for integrated chart--context reasoni…
RSS 官方收录 · 可信分层展示
Monitoring Web Agents Without Internal Signals: Observable Trajectories and Key-Step Supervision
arXiv:2609.02057v1 Announce Type: new Abstract: Reliable web-agent monitoring is difficult when model-internal uncertainty signals such as token logits are unavailable.…
- In this work, we study prefix-level risk prediction for web agents usi…
- We derive two observable trajectory representations: Macro features su…
- Instead of inheriting the final result label, we label the first criti…
RSS 官方收录 · 可信分层展示
HeadWiseKV: Budgeted Per-Head Cache Residency for Hybrid Long-Context Language Models
arXiv:2609.…
- 02029v1 Announce Type: new Abstract: Long-context inference retains a …
- This bottleneck remains in hybrid language models because their residu…
- We study how to allocate this state under an aggregate KV-residency bu…
RSS 官方收录 · 可信分层展示