@Ryrenz: 北大图灵班出身的博士刚开源了一份算法竞赛宝藏 awesome-competitive-olympiad-algorithms,从 NOIP 到 NOI、ACM-ICPC 难度的原创题和讲义全在里面。 作者 Zijian Wang 的 Gi…
摘要
北大图灵班出身的博士 Zijian Wang 开源了 awesome-competitive-olympiad-algorithms,收录其原创的 NOI、省队选拔及 ACM-ICPC 难度题目、C++ 标准解答、中文题解与讲义幻灯片,并附带官方测试数据。
查看缓存全文
缓存时间: 2026/11/20 14:05
🥇 北大图灵班出身的博士刚开源了一份算法竞赛宝藏 awesome-competitive-olympiad-algorithms,从 NOIP 到 NOI、ACM-ICPC 难度的原创题和讲义全在里面。
作者 Zijian Wang 的 GitHub 主页写着:北大图灵班本科、北大前沿计算研究中心博士,2020 年 World Finalist,现在在小马智行。
学竞赛最缺的往往不是题,是出题人的视角:这道题为什么这么设计、正解是怎么想到的。awesome-competitive-olympiad-algorithms 收的是作者自己给 NOI、北京省队选拔和模拟赛出的原创题,每道题配齐三样:题面、C++ 标准解答、带正确性和复杂度分析的题解。
另外还有一批讲义幻灯片,按难度排好了顺序,题目目录也做了无剧透版本。官方测试数据单独打包,Google Drive 和 GitHub Release 都能下,自己写完可以直接对拍。讲义和题解全是中文;代码 MIT,题面和讲义是 CC BY-NC-SA 4.0,不能商用。
跟着出题人的思路刷题,和闷头刷题完全是两种体验。
GitHub:
wzj52501/awesome-competitive-olympiad-algorithms
Source: https://github.com/wzj52501/awesome-competitive-olympiad-algorithms
awesome-competitive-olympiad-algorithms
A personally curated archive of original algorithmic problems and lecture notes — every problem authored and set by hand — spanning NOIP through NOI and ACM-ICPC level. Centered on the algorithmic craft shared by informatics olympiads and ACM-ICPC-style contests: design, analysis, and proof — not low-level implementation.
About
This repository collects materials I created over years of competing and problem setting in algorithm contests — all original work:
- Lecture slides — notes prepared for teaching sessions.
- Original contest problems — problems I set for the National Olympiad in Informatics (NOI), provincial team selection, and mock contests.
Difficulty ranges from NOIP all the way up to NOI and ACM-ICPC level.
Whether you aim for OI, ACM-ICPC, or simply want to study rigorous algorithm design, I hope this archive offers something useful. If it does, a star would mean a lot — and would help more people discover it.
Each problem is preserved with three artifacts:
- statement — the problem description
- standard solution — a reference C++ implementation
- editorial — the algorithm explained, with correctness and complexity
Structure
.
├── Lectures/ Algorithm lecture slides, notes, and a difficulty-ordered guide
└── Setter/ Original problems, grouped by contest, with a spoiler-free guide
├── NOI/ National Olympiad in Informatics
├── BJTSC/ Beijing Team Selection Contest
└── NOIP/ National Olympiad in Informatics in Provinces
| Directory | Contents |
|---|---|
Lectures/ | Lecture slides and notes on algorithms, with a difficulty-ordered guide covering key topics, prerequisites, and recommended levels. |
Setter/ | A spoiler-free guide to the original problem sets, their sources, approximate difficulty, and suggested training order. |
Setter/NOI/ | NOI problems, NOI-level mock contests, and problem-setting or validation archives. |
Setter/BJTSC/ | Beijing Team Selection contests, joint provincial selection materials, and related training rounds. |
Setter/NOIP/ | NOIP-level mock contests. |
Note — All lecture slides and editorials are written in Chinese.
Test Data
The official test data (test points: .in / .out / .ans) for every original
problem is published separately as AwesomeAlgorithms_SetterData, a data mirror
of the Setter/ directory tree — one sub-folder per problem, original file names
kept.
- Download: AwesomeAlgorithms_SetterData.zip (Google Drive)
- Also available from GitHub Release v1.1
The archive includes its own README.md documenting the exact format.
License
- Code (standard solutions /
.cpp): MIT License - Content (problem statements, editorials, lecture slides): CC BY-NC-SA 4.0
Updates
- upd0910: Add difficulty-ordered guides for all lecture materials and a spoiler-free catalog of the problem sets, including prerequisites, intended levels, sources, and suggested study order.
- upd0909.2: Fix typos and minor issues; make the standard solutions C++17-compatible.
- upd0909.1: Release the official test data (AwesomeAlgorithms_SetterData) as GitHub Release v1.1.
- upd0908: Add 12 new lecture slides and the MIT License; complete the missing solutions for BJTSC Mock-1 and NOI Mock-1.
相似文章
@yaojingang: 经好友拔刀刘同意,将他们近期抓取的国内各大AI平台数据集进行开源 清洗后的数据集、论文预印本以及第一份分析报告,已推送到GitHub仓库,地址见评论区 这应该是面向国内各大AI平台的最新、最全的一份公开GEO原始数据集 包括了豆包、Dee…
好友拔刀刘开源了国内8个AI平台的搜索结果数据集,包含620个标准问题、21万条引用记录,并提供清洗后的数据、论文预印本和分析报告。
@DayShuai: 明天 volunteer 了在 Yang Zhang lab 组会上分享我自己的 AI loop。同一套 OS pattern 最近半年在 automath 和 newmath 跑出 3,400+ 0-axiom Lean 4 定理、5×…
分享在Yang Zhang实验室组会上的AI loop经验,包括自动化定理证明、多机器协作、蒸馏私人经验库等,并提及Fields奖得主使用AI解决数学难题的例子。
@hassanyehiaa: 这两天收到很多私信,也有很多家教和老师提出建议,真心十分感谢 我也一直在完善知识图谱内容,并且根据大家的建议对题库的难度系数进行了增加,同时增加了题库的数量 具体如下:难度分档: 每个知识点的题分成基础、提高、拓展三档。 诊断顺序: 先出…
作者更新了开源的中小学知识图谱项目 cn-math-taxonomy,为每个知识点增加基础/提高/拓展三档难度题库,并改进自适应诊断顺序与前置知识回溯逻辑。项目覆盖数学、语文、科学、英语、物理,提供在线诊断工具,但声明为 AI 生成初稿、尚待教师审核。
@FinanceYF5: Google DeepMind发布AlphaProof Nexus论文: AI agent在353个开放数学问题中自主解决了9个Erdős难题,包括两个56年未解问题,并证明了44个OEIS猜想。 每道题推理成本仅数百美元。
Google DeepMind 发布 AlphaProof Nexus 论文,AI agent 在 353 个开放数学问题中自主解决了 9 个 Erdős 难题(包括两个 56 年未解问题),并证明了 44 个 OEIS 猜想,每道题推理成本仅数百美元。
@zhizhuxia22: 我又又整理了一份AI圈优质博主合集 1)@lxfater(铁锤人)AI创业实战,开源项目多 2)@xiaohu(小互)前沿AI动态,讲到你懂 3)@vista8(向阳乔木)实用教程多,工具分享勤 4)@oran_ge(Orange AI)…
这是一篇分享AI圈优质博主和Agent Skills网站合集的帖子,旨在帮助读者找到适合自己的AI学习和工具资源。