Tev: Scaling Long-Horizon Agentic Environments with Dense Progress Rewards· Tev: 使用密集进度奖励扩展长时段代理环境
Long-horizon agents have drawn growing attention for their ability to work autonomously --…
Long-horizon agents have drawn growing attention for their ability to work autonomously --…
Reasoning models achieve strong performance through long chains of thought (CoT), but ofte…
How can we build an agent that thinks beyond human language, yet still acts through it in …
Generating information-rich visual artifacts such as slides, posters, and dashboards as co…
Multi-objective molecular optimization is a critical process in drug discovery that aims t…
Robots are increasingly autonomous in task execution, yet their improvement remains largel…
Large language models (LLMs) generally rely on sampling temperature to regulate the amount…
Pre-trained Large Language Model (LLMs) are effective on many tasks in zero-shot and few-s…
Harness self-evolution revises an agent's software scaffolding, including prompts, memory,…
LLM ensemble gains are often measured against a single baseline, but the value attributed …
Long-horizon agents accumulate progress in persistent task state, yet repeatedly process i…
Large language model multi-agent systems extend individual agents' capabilities through co…
System one models enable direct, low-latency decision-making. Yet recent practice of predi…
Test-time scaling for large language models (LLMs) can involve generating multiple candida…
In this paper, we propose an extremely efficient, training-free method to extract token-le…
Personal-agent memory has largely focused on learning what users know, prefer, and want. H…
Combining pretrained models is a common strategy for building AI systems, whether by align…
An embodied agent must prioritise among competing bodily needs. In active inference, it ca…
Controlling an acting LLM agent means deciding, at each step of a long-horizon task, wheth…
Large language models increasingly tackle tasks that require multi-step reasoning, where e…
Benchmarks report what an Artificial Intelligence (AI) agent can do on a task suite, but n…
Point-in-time (PIT) agents can recall post-cutoff facts even when their external evidence …
Sampling-based test-time scaling enhances LLM reasoning by aggregating multiple reasoning …
System One models choose an agent's next action by scoring a supplied set of options in a …
Automated post-training transforms pre-trained large language models (LLMs) into models ca…
Fast CPU OCR: line detection, CTC recognition, reading order, markdown
The open source AI coding agent, in Rust. CLI, TUI and HTTP server for local AI agents — n…
A desktop app that teaches you how modern AI works by letting you play with it.
We'll likely learn more about how this will work at the company's upcoming "AI & Autonomy …
Dial in the next gaming session this GFN Thursday. New NVIDIA DLSS 4.5 controls give GeFor…
Celebrate Teachers’ Day with AI for Educators in Brazilian Portuguese and see how Study Mo…
Amazon-owned Bee has launched its first new AI devices since its acquisition, including an…
真正的双11于10月20日开卖。10月15日,天猫发布会披露,今年天猫双11吸引了100万全球品牌、7200万款商品参与,均创历史新高。 AI和旗舰店,成为今年双11的两个关键词。…
10月15日,在2026香港秋季电子产品展上,万拿机器人全球首发NOVA21灵巧手。面向机器人科研与开发团队,万拿NOVA21采用21自由度全主动旋转直驱设计,集灵巧、力量、精准与…
近日,基元律动(Token Rhythm)宣布已有超过30万名开发者入驻,同期推出双11预售限时福利。根据官方信息,用户在 TokenRhythm API平台单笔充值满50元,即可…
Much of today’s scientific progress begins with computation. But turning accelerated compu…
10月15日,苹果董事会执行主席蒂姆·库克在上海走访,接受媒体采访。在谈及Apple Intelligence(苹果智能)相关问题时,库克表示苹果智能及相关阿里千问大模型确已获得批…
10月12日-16日,2026年中国国际工业博览会在上海举办。海康威视以“物联感知+AI,加速工业场景数字化”为主题,携面向质量检测、设备监测、生产运营、安全生产等核心环节的100…
随着新能源汽车和半导体产业持续增长,一个关键环节正被愈加关注:零部件质检。检测标准日趋严苛,传统依赖人工经验和抽检的模式,逐渐成为产品质量、产能爬坡的隐性瓶颈,而加持了AI的超声质…
<section data-mpa-powered-by="yiban.io" data-pm-slice="0 0 []" style="-webkit-tap-highligh…
上海人工智能实验室(上海AI实验室)访问学者、大湾区大学教授许地生团队依托书生科学多模态大模型S2(书生-S2),基于“通专融合”技术路径,通过人机协同完成了三维Ghys猜想证明。…
机器人数据采集正在经历一条“解绑本体”的演进路线:从遥操作到 UMI再到 Ego,设备越来越轻,采集行为越来越简单。 不过,在不断简化设备的过程中,一些关键交互信息也随之流失——E…
<s
10月15日,百度官方公众号发文公布百度世界2026五大亮点,涵盖李彦宏年度主题演讲、百度全栈AI升级、智能组织进化等内容。 据悉,百度世界2026将于11月12日在北京国家会议中…
冷汗时刻
网友集体崩溃:新模型要来了?
给企业Agent打了个样
The founders detailed how their startup is developing AI that can design chips, learn from…
Every day, conversations unfold across social media and websites around the world. A seemi…
The label will appear on the player page and include songs entirely generated by AI or par…
Most companies use AI agents to ship products faster or speed up workflows. Apate.ai sends…
<
Deezer said that the new toggle will be rolled out by the end of the year.
Get 50% discount on the last two days of TechCrunch Disrupt 2026, Oct. 14-25, in San Franc…
Zenzap is used by organizations including Starbucks and Taco Bell.
The problem Floqer is trying to solve is familiar to anyone who has worked in sales or rev…
Day 2 at TechCrunch Disrupt 2026, full of all themes across AI, investor-only session, mor…
See how Amp pushes the frontier of what agents can do with GPT-6 Astra, Ultrafast, GPT-Liv…
With OpenAI Daybreak, SentinelOne cuts weeks of malware analysis to hours, creating capaci…
Block is building AI-powered security with OpenAI. Its security team investigates threats …
With GPT-6 Astra and GPT-6.1 Sol, Codex, and Daybreak for cybersecurity, this AI lab build…

YouTube · OpenAI

YouTube · NVIDIA

YouTube · DeepLearningAI

YouTube · OpenAI

YouTube · Google for Developers

YouTube · NVIDIA