We are publishing a list of QOJ problems for which LLMs, mostly GPT, have found solutions that are significantly faster than the authors’ intended solutions. If you would like to submit additional problems for inclusion in this list, please write an editorial on QOJ in the problem forum, mark it as anonymous, and leave a comment below.
For some of the tasks, LLMs even found mistakes in author's solution. Those tasks are marked with †.
| Task | From | Author's Solution | GPT's Solution | Article |
|---|---|---|---|---|
| Make It Divisible | ICPC Asia Hangzhou Regional 2024 | $O(T \sqrt {V} + n (\log n + \log V) + n\cdot d(V))$ | $O(n + V^{1/3} \log V)$ | by GPT-6 Pro |
| †Dreamy Putata | ICPC Asia Hangzhou Regional 2023 | $O(qm^3 \log n)$ | $O(qm^2 \log n)$ | by GPT-6 Pro |
| 机器人 | Chinese NOI 2019, Day 1 | $O(n^3)$ | $O(n^2 \log n)$ | by GPT-6 Pro |
| 序列变换 | Chinese NOI 2025, Day 1 | $O(N^2)$ | $O(N + \log P)$ | by GPT-6 Pro |
| 彩虹树 | Chinese NOI 2026, Day 2 | $O(n^4)$, $O(n^3 \log n)$ | $O(n^3)$ | by GPT-6 Pro |
| Not a work of Idol | CCPC Finals 2024 | $O(4^p \cdot \text{poly} (p) + T p^2 \log_p^2 n)$ | $O(p^5 \binom{3q}{q} + p^6 + T p^2 \log_p^2 n) < O(2.6^p + T p^2 \log_p^2 n)$ | by GPT-6 Pro |
| Under the Epilogue | CCPC Finals 2024 | $O(n^6)$ | $O(n^{\omega(1,2,1) + \varepsilon}) < O(n^{3.251})$ | by GPT-6 Pro |
| Four Kubic Theorem | CCPC Finals 2025 | $O(\sqrt p \log p)$ | $O(\log p)$, $O(\sqrt p)$ | |
| Spirited Away | CCPC Finals 2025 | $O(n \log^2 V)$ | $O(n \log V)$ | by GPT-6 Pro |
| DFS Order 4 | ICPC Asia EC-Final 2023 | $O(n^3)$ | $O(n^2 \log n)$ | by GPT-6 Pro |
| Coloring | ICPC Asia EC-Final 2022 | $O(n^2)$ | $O(n \log n)$ | by GPT-6 Pro |
| 被 EI 加 0 了 | by Elegia | $O(n^3)$ | $O(n^2 \log^2 n)$ | by GPT-6 Pro |
| 欧拉？欧拉！ | CTT 2022 Day 4 | $O(n^5)$ | $O(n^2 \log^2 n)$ | by GPT-6 Pro |
| 旧试题 | SDOI 2018, Round 2, Day 2 | $O(n^{1.5} \log^3 n)$ | $O(n^{1+o(1)})$ | by GPT-6 Pro |
| 工业系统 | NOI 2026 China Multi-Provincial Selection | $O(n\sqrt{n+m})$ | $O(n \log n + m\log^2 n)$ | by GPT-6 Pro |
| 最大值 | by Elegia | $O(n \cdot k^{1.5})$ | $O\left(\min\{n^2k,\ n^{11/2}\sqrt k\log(nk)\}\right)$ | by GPT-6 Pro |