AksLolCoding's blog

By AksLolCoding, history, 41 hour(s) ago, In English

DuyNgaDocTon1410 is an AI cheater.

Firstly, let's look at their performance. In his first Div. 2 round, they place 22nd. In the next one, they places 2nd (as a pupil), solving every single question! The same day, they take another Div. 2 contest, and fail to even solve two questions, ending at to 6690th place, despite attempting B, C, and D multiple times. Remember, they just got 2nd place a few hours earlier. It is not possible for a human to vary this much in performance on the same day.

Now, lets look at their submissions. One thing immediately caught my attention: out of his 100 most recent submissions, 33 of them are WA on (pre)test 1, and 5 are compilation errors. So this person doesn't test their code before submitting, and doesn't even check if it compiles. Very common among AI copy-pasters.

If we look at the code in these submissions, it is not immediately clear that he uses AI. However, there are too many signs of AI for this to be a coincidence. For example, most of his submissions have no comments, while others such as 310106426 have some of the most detailed and perfect comments I have ever seen. Another discrepancy is his header usage: some of his solutions use bits/stdc++.h, while others include each header individually, and one of his solutions even uses both (which is completely useless). There is even more evidence when you look at his coding style.

Altogether, DuyNgaDocTon1410 is obviously using AI to cheat. Nothing else can explain this amount of evidence. Please ban him and all other AI cheaters so that Codeforces remains fair.

Unrelated: while looking at several of his submissions my account got flagged as being used to crawl, but I was just browsing through his AI submissions.

Full text and comments »

  • Vote: I like it
  • +54
  • Vote: I do not like it

By AksLolCoding, history, 3 days ago, In English

For implementations, see my submissions

A — Serval and String Theory

Solution

B — Serval and Final MEX

Solution

C — Serval and The Formula

Solution

D — Serval and Kaitenzushi Buffet

Solution

E — Serval and Modulo

Solution

F1 — Serval and Colorful Array (Easy Version)

Solution

F2 — Serval and Colorful Array (Hard Version)

Solution

Overall, contest was fun and all problems were very good.

Full text and comments »

  • Vote: I like it
  • +28
  • Vote: I do not like it

By AksLolCoding, history, 6 days ago, In English

Let's turn the controversy up to 11 (not in binary). Feel free to post some more polls in the comments if you want to

What is this data structure called?

Disjoint Set Union

Union-Find

How do you indent?

2 Spaces

3 spaces

4 Spaces

Tabs

No indentation (you need help)

Best language for CP?

C

C++

Python

Java

What type of variable do you prefer?

Local

Global

How do you bfs/dfs?

Deque

Recursion

Deque for bfs, recursion for dfs

Some from the comments:

(C/C++ only) How do you I/O?

Cin/cout

Printf/scanf

Custom superfast io

What range do you use?

inclusive-inclusive [l,r]

inclusive-exclusive

exclusive-exclusie (l,r)

Segment Tree

Array based

Pointer based

Recursive

Iterative

Indexing

0-index

1-index

Modulo

1000000007

998244353

Debugging

print debugging

debugger

Lambda or function?

lambda

function

(Inspired by neal.fun)

Full text and comments »

  • Vote: I like it
  • +63
  • Vote: I do not like it

By AksLolCoding, history, 3 weeks ago, In English
  • Vote: I like it
  • +2
  • Vote: I do not like it

By AksLolCoding, history, 3 weeks ago, In English

I have a problem as follows: You are given a weighted tree and a number L. You can do the following exactly once: choose a pair of nodes and add an edge of length L between them. Over all possible resulting graphs, what is the smallest possible diameter?

The editorial states that it is optimal for the chosen pair of nodes to both lie on a diameter of the original tree, but the proof is left as an exercise to the reader. I could not prove this, but can anyone explain why this is always optimal?

Full text and comments »

  • Vote: I like it
  • +1
  • Vote: I do not like it

By AksLolCoding, history, 6 weeks ago, In English

Jiangly's profile now shows the title of "Rating4000_887306" at the top but still says "jiangly" everywhere else.

Full text and comments »

  • Vote: I like it
  • +26
  • Vote: I do not like it