1
卡普空开源 RE:Dox 结构化数据引擎，JSON 反序列化最高提速 2.84 倍
IT之家 10 月 5 日消息，卡普空宣布，将旗下最新的数据处理及序列化基础库“RE:Dox”以开源软件（OSS）形式公开。该项目是卡普空次世代游戏引擎计划 REX（RE neXt ENGINE）的基础技术之一，开发者可以自由使用。 卡普空早在 2023 年 10 月便公布了 REX 引擎计划，目标…
2
清华稳准智能联合发布LimiX-2，结构化数据基础模型登顶国际评测榜单
9月16日，稳准智能联合清华大学发布新一代数据大模型 LimiX-2，模型参数规模提升至400M。
3
Enhancing LLMs in Predictive Political QA with Semi-Structured Data
半结构化数据让大模型预测政治走向更靠谱？看这项研究如何提升问答准确性。
arXiv:2608.21218v1 Announce Type: new Abstract: Predictive political question answering (QA), such as predicting how a political actor will vote, goes…
4
Training Fair Tabular Foundation Models
表格数据基础模型如何兼顾公平性？ICML spotlight论文提出全新训练思路，AI公平性研究者必读。
arXiv:2608.14211v1 Announce Type: cross Abstract: Tabular Foundation Models (TFMs) have emerged as leading methods for tabular predictive tasks, lever…
5
Show HN: Emboss – a Python PDF engine that keeps documents as structured data
一个用语义模型保持文档结构化、自动生成无障碍PDF的Python引擎，EU可访问法案下必备。
Article URL: https://github.com/GGChamp85/Emboss Comments URL: https://news.ycombinator.com/item?id=49106579 Points: 5 # Comments: 1
6
I built an eval harness to prove an LLM worked. It proved the opposite.
通过构建评估基准，揭示LLM在结构化数据提取上与正则表达式高度一致，提醒我们警惕AI万能论。
I set out to have a language model classify integration failures. I built an evaluation harness to prove it worked. The harness proved it wasn't worth…
7
How to Build Rich Link Previews Without Maintaining a Browser Instance
通过API快速获取链接的OG元数据、截图和结构化数据，无需维护浏览器实例，解决内存泄漏和路径解析问题。
The problem with link previews You paste a URL in Slack and get a nice preview card. Title, description, maybe a thumbnail. Users expect the same thin…
8
Show HN: Aict – Unix coreutils that output XML/JSON, built for AI agents
专为AI代理设计的Unix coreutils替代品，输出XML或JSON格式，让命令行工具无缝对接智能体工作流
Article URL: https://github.com/synseqack/aict Comments URL: https://news.ycombinator.com/item?id=48920681 Points: 2 # Comments: 0
9
DKCD: Domain Knowledge-Enhanced Causal Discovery from Unstructured Data
如何从非结构化数据中发现因果？本文提出利用领域知识增强因果发现的新方法，突破传统局限。
arXiv:2607.09348v1 Announce Type: new Abstract: Causal discovery from unstructured data is a challenging yet underexplored task in high-expertise doma…
10
I built a tool to check what AI agents actually understand about your website
一个检测AI代理实际理解你网站到何种程度的工具，帮你填补声明与真实数据间的鸿沟。
There's a gap between what a website says about itself and what an AI agent can actually verify from its structured data. I got curious about this aft…
11
Show HN: BrainAPI's event-centric graph from any data with four-stage pipeline
将任意数据转化为事件中心图，通过四阶段管道精准捕捉实体关系，轻松回答多步复杂问题。
Article URL: https://github.com/Lumen-Labs/brainapi2 Comments URL: https://news.ycombinator.com/item?id=48676665 Points: 2 # Comments: 0
12
Random Rule Forest (RRF): Interpretable and Manageable Ensembles of LLM-Generated Questions for Predicting Success from Unstructured Data
用LLM生成问题构建可解释的随机规则森林，从非结构化数据中预测成功，兼顾性能与可管理性。
arXiv:2505.24622v3 Announce Type: replace Abstract: Many high-stakes screening tasks require predicting rare outcomes from unstructured text, where er…
13
Strengthening LLMs for Tabular Prediction with Structural Priors
用结构先验强化大模型处理表格数据预测，一项提升LLM表格推理能力的创新方法。
arXiv:2510.17385v5 Announce Type: replace-cross Abstract: Tabular prediction has long been dominated by gradient-boosted decision trees and specialize…
14
Show HN: Tracore – We turned our resume parser into a document-to-JSON API
把简历解析器变成通用文档转JSON API，一步定义 schema，API 上传、webhook 接收结构化数据，欧洲 GDPR 合规。
Article URL: https://tracore.io/en/ Comments URL: https://news.ycombinator.com/item?id=48551491 Points: 1 # Comments: 0
15
Categorical Prior Lock-in: Why In-Context Learning Fails for Structured Data
揭秘大模型在结构化数据上的“卡壳”瞬间：类别先验锁定如何让上下文学习失效
arXiv:2606.11961v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as conditional generators for structured data, re…
16
Tabstack Structured Extraction
从网页一键提取数据到结构化JSON，无需写爬虫，高效搞定数据采集。
Extract web data into structured JSON, no scraper required. Discussion | Link
17
AgriGov: A Structured Multilingual Dataset Curation for Indian Government Schemes for Farmers
为印度农民政府计划打造的三语（英-印地-马拉地）结构化数据集，助力农业政策分析与NLP研究。
arXiv:2606.08272v1 Announce Type: new Abstract: AgriGov is a curated, trilingual (English-Hindi-Marathi) dataset designed to address the scarcity of d…
18
When Regex Fails: Using LLMs to Extract Structured Data from Messy Pages
正则表达式搞不定的脏数据，让LLM来当爬虫神器，2-3个示例就能搞定HTML到JSON的提取！
I’ve been doing web scraping for years. For most projects, I lean on BeautifulSoup, cssselect, and a handful of regex patterns. You know the drill: in…
19
DTBench: A Synthetic Benchmark for Document-to-Table Extraction
首个专门评估LLM从文档提取结构化表格的合成基准，揭示现有模型在精确表格生成上的不足。
arXiv:2602.13812v3 Announce Type: replace-cross Abstract: Document-to-table (Doc2Table) extraction derives structured tables from unstructured documen…
20
Ask HN: Does your website show up when ChatGPT recommends tools in your field?
你的网站在ChatGPT推荐中隐身了吗？调查显示80%的SaaS站点毫无存在感，一个叫Agentable的工具正在帮你自查。
AI agents like ChatGPT now do 50M+ product recommendations per day. I've been scanning SaaS sites to see if they show up when users ask AI for tool re…