A Formal Limitation on Learning Human Language From Textual Corpora· 从文本语料库学习人类语言的正式限制
Can a listener recover what a speaker means from the form of an utterance alone? We answer…
Can a listener recover what a speaker means from the form of an utterance alone? We answer…
Modern agent systems assemble capabilities at runtime, and this dynamic composition has re…
Recent advances in generative AI allow users to create 3D models from text or images. Howe…
We investigate whether automatic speech recognition (ASR) errors in user input can lead to…
When training Mixture-of-Experts (MoE) language models with expert parallelism, all-to-all…
Forced alignment evaluation typically requires manually annotated timestamps, limiting lar…
Two forms of test-time scaling for Large Language Models (LLMs) have emerged as effective …
Predicting robot videos requires both precise motion reasoning and preservation of high-fr…
Recent advances in large language models (LLMs) have demonstrated strong capabilities in n…
Factual question answering (QA) typically assumes a single canonical answer, obscuring whe…
When a multi-agent system changes its team, it also changes the messages and final answer …
Current large language models (LLMs) increasingly benefit from external tool integration, …
Proof assistants such as Lean 4 promise the paradigm of formally verified mathematics, but…
Post training a language model to reason means updating its weights. Supervised finetuning…
Across audit applications, judgments must be supported by reasonable evidence. However, st…
In financial markets, a sequential policy that reacts systematically to price movements ma…
Repurchase recommenders in e-commerce are commonly framed as a binary question asking "wil…
We introduce MAP, the first benchmark to evaluate multimodal AI systems as assistants for …
LLM agents increasingly modify their own prompts, tools, middleware, resources, and execut…
Machine Unlearning methods for Large Language Models typically assume pre-specified forget…
Large language models have facilitated knowledge graph (KG) construction from clinical gui…
AGENT-O is a modular ontology framework that defines a semantic Agent Card for representin…
Encoding temporal order is a fundamental requirement for sequence representations in Hyper…
The ever-expanding volume of the chemical literature offers unprecedented opportunities to…
Reservoir Computing (RC) designs Recurrent Neural Networks around a fixed, i.e., untrained…
Ultimate LLM Gym Trainer Guide 2026: Reinforcement Learning Fine-Tuning Framework
AI Form Checker 2026: Perfect Your Squat, Deadlift & Bench Press Form
AI Studio Trainer Hub 2026: Open Source Installer & Auto-Updater Suite
AI视频生成速度超越播放速度,推动实时商业化。
<img class="rich_pages wxw-img" data-aistatus="1" data-croporisrc="https://mmbiz.qpic.cn/s…
8月31日,内容创作平台小云雀AI宣布品牌升级,围绕“一起创作好故事”这一全新定位,升级产品能力、创作者扶持政策和内容生态。未来3年,小云雀AI将投入一亿积分,依托创作者计划、剧本…
全球消费电子产业的“风向标”——德国柏林国际消费电子展(IFA 2026)将于9月4日在柏林展览中心启幕。智能户外庭院科技企业耐士劳,将以双展台联动形式,面向全球客户与合作伙伴,带…
<section data-mpa-powered-by="yiban.io" data-pm-slice="0 0 []" style="-webkit-tap-highligh…
<section data-mpa-powered-by="yiban.io" data-pm-slice="0 0 []" style="-webkit-tap-highligh…
<section data-mpa-powered-by="yiban.io" data-pm-slice="0 0 []" style="-webkit-tap-highligh…
<img data-src="https://static.leiphone.com/uploads/new/images/20260901/6a968ecc57a78.jpg" …
<img class="rich_pages wxw-img" data-aistatus="1" data-croporisrc="https://mmbiz.qpic.cn/s…
<img data-src="https://static.leiphone.com/uploads/new/images/20260901/6a968ddf001b1.jpg" …
<img data-src="https://static.leiphone.com/uploads/new/images/20260901/6a968db139935.jpg" …
<img class="rich_pages wxw-img" data-aistatus="1" data-croporisrc="https://mmbiz.qpic.cn/s…
<img data-src="https://static.leiphone.com/uploads/new/images/20260901/6a968cdb5ccb1.jpg" …
<img class="rich_pages wxw-img" data-ratio="0.55" data-s="300,640" data-type="jpeg" data-w…
美观且支持实时更新的架构图,也能用AI一键生成了!
主打一个明升暗降,理直气壮
近日,VAST宣布完成B轮和B+轮融资,合计约30亿人民币,由经纬创投领投,完美世界、蓝色光标、芯动能投资、延趣游戏、中科创达、广电投资、三七互娱等一线产业资本,鼎晖VGC、中金资…
近日, Violoop 正式完成亿元级天使轮及 Pre-A 轮融资。联想创投、中金保时捷、蓝驰创投、元生资本、启赋资本和 Zero2IPO Ventures</spa
01 存算一体,进入交付时刻 2026年,存算一体赛道的节奏骤然加快。 8月24日,小米发布玄戒O100高带宽AI加速芯片,采用6nm工艺与3D晶圆级堆叠,近存计算带宽达1.22T…
参数冻结,能力暴涨
8 月中旬,新加坡的一场公开演讲上,南洋理工大学副教授、Ropedia 首席科学家刘子纬抛出了一个问题:为什么读过整个互联网的 AI,进了物理世界,却连一杯咖啡都冲不好? 他的回答…
8月28日晚间,国产GPU企业壁仞科技(06082.HK)披露2026年中期业绩。2026年上半年,壁仞科技实现营业收入12.36亿元,较上年同期增长约1997.6%;期内亏损大幅…
连发六篇Nature期刊,复旦Neolab打通全尺度生命推演
<img data-src="https://static.leiphone.com/uploads/new/images/20260901/6a963e69bf6dd.jpg" …
<img data-src="https://static.leiphone.com/uploads/new/images/20260901/6a963df127dfe.jpg" …
一张显卡,开始被越来越多创作者当成生产资料重新算账。 过去,要把AI视频真正搬到本地,首先卡在模型。效果更好的视频模型大多通过云端服务提供,能够下载到本地的开源模型,距离稳定生产往…
<section data-mpa-powered-by="yiban.io" data-pm-slice="0 0 []" style="-webkit-tap-highligh…
<img class="rich_pages wxw-img" data-aistatus="1" data-croporisrc="https://mmbiz.qpic.cn/s…
<section data-mpa-powered-by="yiban.io" data-pm-slice="0 0 []" style="-webkit-tap-highligh…
Apple says it has evidence that a former employee destroyed evidence of data theft after l…
Versions of OpenAI's ChatGPT and SpaceXAI's Grok will join Google's Gemini on the Pentagon…
As frustration over AI influencers has been growing, Instagram is limiting the reach of un…
The three-year-old startup says it reached $15 million in ARR and profitability before rai…
Nvidia invests $3.5 billion into Taiwanese chipmaker MediaTek. The deal shows how Nvidia p…
划重点:这次不是你挨家敲门,是VC排队来看你。
Polimill uses OpenAI GPT models and Codex to help municipalities search and use administra…
OpenAI supports California SB 1119, advancing strong, age-appropriate AI safeguards for te…
Investors, founders, and operators from across Europe arrived for the annual Nordic TechBB…

YouTube · Two Minute Papers