1
国家数据局：全国已建成高质量数据集超 12.6 万个，总体量超 1815PB
IT之家 8 月 29 日消息，据央视新闻昨日报道，2026 中国国际大数据产业博览会正在贵州贵阳举行。国家数据局相关负责人表示， 全国已建成高质量数据集 超 12.6 万个 ，为人工智能发展提供支撑。 高质量数据集，是指经过采集、加工等数据处理，可直接用于开发和训练人工智能模型、能有效提升模型性能…
2
Hugging Face reportedly in talks to be acquired for $13B
AI模型社区与托管平台，汇聚海量开源模型和数据集，支持在线推理与微调，开发者必备。
Hugging Face has reportedly been fielding acquisition offers that would value the company at around $13B. But with the founders' feeling of responsibi…
3
AGIDefect-4K: A Richly Annotated Dataset for AI-Generated Image Defect Detection, Localization and Explanation
首个4K级AI生成图像缺陷检测数据集，支持检测、定位与解释三合一，是图像生成质量评估的利器。
arXiv:2608.20713v1 Announce Type: new Abstract: Generative AI can now produce highly realistic images, yet current models still exhibit subtle but cri…
4
诺亦腾机器人发布 HiPHI，开源 617.5 小时高精度人体运动数据
开源617.5小时高精度人体运动数据，覆盖全身动作与人-物交互，助力人形机器人和数字人研究
IT之家 8 月 23 日消息，诺亦腾机器人在 2026 世界机器人大会期间发布 HiPHI，这是一套面向人形机器人学习、数字人，以及计算机图形学领域研究人员和工程师的高精度光学动作捕捉数据集。 据报道， 该数据集总长 617.5 小时 ，内含 371.8 小时的全身人体运动数据，以及 245.7 …
5
PandasCorpus: A Resource of Real-World Pandas Workflows and Usage Patterns
超大规模真实世界 Pandas 语料库，4M+ API 调用揭示开发者实战习惯，数据科学工具链优化必读。
arXiv:2608.14742v1 Announce Type: cross Abstract: Pandas has emerged as the de facto library for data processing and machine learning, widely used for…
6
LAPF: LLM-Agent-Based Path Finder Using the UAVScenes Dataset
用大模型Agent驱动无人机路径规划，UAVScenes数据集加持，AI落地 robotics 的新思路。
arXiv:2608.15175v1 Announce Type: cross Abstract: Uncrewed aerial vehicles (UAVs) are increasingly deployed for autonomous navigation in complex outdo…
7
RamseyGadgets: A Graph Construction Dataset for LLMs
LLM能像人类一样构造特殊图吗？这个数据集专门来测，为图论+AI交叉研究提供新基准。
arXiv:2608.14999v1 Announce Type: cross Abstract: Constructing special graphs is an important task within graph theory and computer science. Many popu…
8
A Survey of Large Models in Sports
全面梳理大模型在赛事分析、运动表现与体育媒体中的前沿应用，附数据集与挑战总结。
arXiv:2608.14377v1 Announce Type: new Abstract: Sports have witnessed growing global enthusiasm in recent years, serving as a vital force for physical…
9
PhyCheck: Fine-Grained Evidence-Grounded Dataset for Physical Law Understanding in Video-LLMs
给视频大模型“补物理课”的新数据集，用细粒度证据检验模型对物理定律的理解力。
arXiv:2608.02150v1 Announce Type: cross Abstract: Embodied intelligence and world models require video understanding systems to go beyond recognizing …
10
Rethinking PPG-based Sleep Staging: Datasets, Metrics, and Benchmarks
可穿戴PPG睡眠分期研究迎来新基准，一文厘清数据集与评估指标，是睡眠监测技术落地的关键参考。
arXiv:2608.00943v1 Announce Type: cross Abstract: Automated sleep staging assigns discrete stage labels to successive time epochs throughout an overni…
11
A Benchmark Dataset for MLLM-Generated Image Detection: GPT Image2 & Nano Banana2
专测AI生成图片的基准数据集出炉，聚焦GPT Image2与Nano Banana2，为图像真伪鉴别提供新标尺。
arXiv:2608.01258v1 Announce Type: new Abstract: The realism of images generated by multimodal large language models (MLLMs), such as GPT Image2 and Na…
12
TIDES: A Longitudinal Bilingual Dataset for Modeling Multi-Party Social Dynamics
纵向双语多方社交数据集，为建模真实对话中的社会动态提供全新基准。
arXiv:2608.01724v1 Announce Type: cross Abstract: Group conversations are fundamental to human collaboration, yet standard large language models (LLMs…
13
The TomeVault Instruction Corpus (2026-07)
大规模指令语料库发布，含近23万文件，可自由用于模型训练与评测。
The TomeVault Instruction Corpus De-identified structural measurements of AI instruction files (CLAUDE.md, AGENTS.md, SKILL.md, .cursorrules and relat…
14
HSS-Synth: Humanities and Social Sciences Data Synthesis for LLMs
人文社科数据稀缺？这项研究提出HSS-Synth合成方案，专攻开放场景下的大模型数据难题。
arXiv:2607.27379v1 Announce Type: new Abstract: High-quality, diverse data are vital for large language models (LLMs) but remain scarce and costly. Da…
15
Yelp's OpenAI Agreement Sets Up ChatGPT Local Discovery With Reviews and Business Data
Yelp与OpenAI达成数据合作，让ChatGPT能基于真实点评做本地推荐，一场搜索引擎替代战悄然打响。
Yelp has confirmed an agreement with OpenAI to extend its trusted content into the AI ecosystem, creating a foundation for AI-driven local discovery o…
16
What is Good? Extracting and Testing Implicit Theories of Literary Quality from LLM Reasoning Traces
用DeepSeek推理轨迹提取文学质量的隐含理论，分类79.3%准确率，揭示AI如何理解好作品
arXiv:2607.20425v1 Announce Type: new Abstract: What makes writing "good" remains a persistent question in literary studies and computational linguist…
17
Ask HN: How do I reliably eval my AI models
AI模型评估没头绪？这篇HN帖求助如何可靠评估AI模型，并讨论寻找专家制作金标准数据集的必要性与渠道。
I created an AI agent and have been considering using Braintrust or Arize - but where can I source experts who actually can make goldsets for me too? …
18
国家数据局：全国已建成高质量数据集 12 万个、总体量超 1565 PB，为 AI 发展提供有力支撑
国家数据局最新数据：全国已建成12万个高质量数据集，总量超1565PB，为AI发展提供有力支撑，并透露七个数据标注城市进展。
IT之家 7 月 19 日消息，据央视新闻从国家数据局了解到，我国持续加大优质数据供给， 已建成高质量数据集 12 万个。 IT之家获悉，截至今年 6 月底，全国已建成科学研究、工业制造、医疗卫生、教育教学等行业领域的高质量数据集 总体量超 1565 拍字节 （PB、1PB=1024TB）， 较一季…
19
国家数据局：全国已建成高质量数据集12万个
全国已建成12万个高质量数据集，数据资源建设取得阶段性成果
记者今天从国家数据局了解到，我国持续加大优质数据供给，已建成高质量数据集12万个。截至今年6月底，全国已建成科学研究、工业制造、医疗卫生、教育教学等行业领域的高质量数据集总体量超1565拍字节（PB），较一季度末增长超60%，现有体量相当于中国国家图书馆数字资源总量的547倍左右，为人工智能发展提供…
20
LessonBench-V1: A Benchmark Dataset for Evaluating AI Lesson Generation Agents
首个专为评估AI课程生成能力设计的基准数据集，融合多种教学模型，提供3620个教学目标的完整评估框架。
arXiv:2607.13041v1 Announce Type: cross Abstract: Large Language Model (LLM) based AI educational content generation systems are increasingly being de…