I was going through the standings of last div 2 contest. No need to tell that the first three (the_chill, e4rendel, nguyenphong233) are cold-blooded cheaters.

The interesting part is the third person (nguyenphong233). Looking at his previous contests I feel like he was solving on his own before. But last night he probably got so frustrated that he bought the paid version and AC'd all the problems. The first two seem like habitual cheaters already but it is honestly sad that the third guy ended up joining them. Of course there are countless others exactly like him who keep shifting from "using own brain group" to "brain-dead" group.




If you think Codeforces has become too annoying because of these brain-dead cheaters, just take a break. Dont become one of them. Otherwise soon you also can't differentiate between what's right and what's wrong just like them.

Just noticed this and wanted to share it here. Good day.

Today in Codeforces round div 2 if you look at the top standings most of them are cheaters (could be all top 20). For example : Daksh-ugi just look at his code. All comments of LLM and compilation errors on test 1. Ofcourse I mean he knows that he will get away with this and sleep another night (i dont know how knowing what a loser he is).

That's not the thing. The thing is I think codeforces could do manual checking for the top players at least for the ones who solves the full set. This will take i dont know like 10 minutes? But will definitely remove a lot of these brain dead people. I am just curious does anyone who can disable the cheaters does not look at the standings even once? I looked for like 2 minutes and found quite a few cheaters. So even 5 to 6 minutes could do a huge work and remove so many of them. Or better if you remove all the newbies and unrated players from the top 100 by default. I dont know why this is not done. For this Codeforces is gonna die soon and only offline contests will prevail.

Also it sometimes saddens me to live in a world most of these people just do this (cheating) for nothing. They just do so. I feel so sorry for them. I mean i wonder what type of people they actually are in real life. Cuz there are actually a lot of them. Okay sorry for this. I just wanted to share this here. MikeMirzayanov sir please take a look into it.

Good day people. Good day.

Distinct Sums Grid
I tried then searched for the solution online but could not find it anywhere. If anybody has solved it please do share the solution here or in private. Thanks in advance.

A huge shoutout to the Div2 winner JuanSign. He was facing many issues till Hello 2026. He was at a newbie level. But for one month he practiced so hard that he became a symbol of inspiration. Inspiration for those who thinks that it takes years of practice to become GMs or LGMs. He managed to do so in less than a month. Today after only three contests (changing his coding style) he managed to surpass Masters, GMs and LGMs with an astonishing win. You have proved that with 20 days of hard practice even this can happen. A huge round of applause for this iconic coder.

Problem Link : Third Permutation
Here how is this code not giving a TLE ?

I got this code from an LLM (i modified it afterwards) but the reason why this is working is not cleared any how. If anyone has solved this in any other way plz do share. Thanks in advance.

    int n;
    cin >> n;
    vector<int> a(n), b(n);
    for (int i = 0; i < n; i++) {
        cin >> a[i];
    }
    for (int i = 0; i < n; i++) {
        cin >> b[i];
    }
    if (n == 2) {
        cout << "IMPOSSIBLE\n";
        return;
    }
    vector<int> c(n);
    for (int i = 0; i < n; i++) {
        c[i] = a[(i + 1) % n];
    }
    for (int i = 0; i < n; i++) {
        label: 
        if (c[i] == b[i]) {
            for(int j = 0; j < n; j++) {
                if(c[i] != a[j] and c[i] != b[j] and c[j] != a[i] and c[j] != b[i]) {
                    swap(c[i], c[j]);
                    goto label;
                }
            }
            cout << "IMPOSSIBLE\n";
            return;
        }
    }
    
    for (int i = 0; i < n; i++) {
        cout << c[i] << " ";
    }

Problem Link
Has any one solved this problem recently? All the expected value problems in Mathematics section seems to have weird test cases. It is very irritating. No matter how i handle the precision it is getting wrong ans in some test cases everytime for the last digit in the answer. If any one has solved it in recent time (mayb more test cases were added) please do share your code here. Thanks in advance.

Problem link

Can someone tell me how to solve this ? I have thought about it for a while but could not draw any conclusions. Thanks.

I am fed up with cf problems. They are sometimes too hard for me. So i am thinking of taking a break from here. I used to do CSES and solved a lot of them and now i am thinking of getting back to it.
My question is at least what rating can i expect after solving all the CSES problems (ofcourse excluding the additional problems and advanced techniques). Please if any one can give me an idea about that then i will do it wholeheartedly from now on.

More specifically what i am trying to ask is if you have done a lot of CSES problems did they actually help you in improving in cf or in doing harder problems in general ? If Yes, pls share your thoughts. Otherwise i dont think i should focus on CSES and maybe shift to something else.

Please don't downvote for no reason. You can simply ignore if you don't like it. Thanks in advance.

Problem link
Hello people. I cant understand how the first player is winning on test case 1. I saw the solution used xor of nim game. But my way of thinking was counting the total number of moves in the game and decide winner based on that. But this fails on the sanple test case. Am i missing something? or is the question statement wrong.

 int n;
    cin >> n;
    int c = 0;
    for (int i = 1; i <= n; i++) {
        int x;
        cin >> x;
        if (i > 1) c += (x * (i - 1));
    }
    if (c % 2)
        cout << "First\n";
    else
        cout << "Second\n";

Problem link

I am trying to run dfs and storing the parent of a child whenever i am finding a longer path to it by using depth but it is giving wrong answers on four test cases which are quiet big to test. I came to know about the toposort and dp solution but I don't understand why doing it this way is wrong. Can anyone tell why this approach is failing or any test case where this approach will fail ?

void dfs(int i, vector<vector<int>> &g, vector<bool> &vis, vector<int> &depth,
         vector<int> &par) {
    vis[i] = 1;
    for (auto child : g[i]) {
        if (depth[child] < depth[i] + 1) {
            depth[child] = depth[i] + 1;
            par[child] = i;
        }
        if (!vis[child]) dfs(child, g, vis, depth, par);
    }
}


vector<int> depth(n, 0), par(n, -1);
    vector<bool> vis(n, 0);
    dfs(0, g, vis, depth, par);
    if (par[n - 1] == -1)
        cout << "IMPOSSIBLE\n" << endl;
    else {
        int x = n &mdash; 1;
        vector<int> ans;
        while (x != -1) {
            ans.push_back(x);
            x = par[x];
        }
        reverse(ans.begin(), ans.end());
        cout << ans.size() << endl;
        for(auto i : ans) {
            cout << i + 1 << ' '; 
        }
        cout << endl;
}

Problem link

Hi there. Can anyone share how to solve this problem? I think it can be solved using Manacher algo but i just can't figure out how to do this optimally. If anyone has solved this using manacher or any other way please do share your idea here. I have tried for a lot of time but failed.

Thanks in advance.

Big-time congratulations to India’s first ever LGM Dominater069! He achieved the LGM title in a remarkably short span of time. Many of us (myself included) are curious about his practice strategy and how much time he has dedicated to cp each day.

So, a humble request to Dominater069: please sir consider writing a blog about your journey whenever you have the time. And if you prefer not to write a full blog, maybe you can leave a comment summarizing your approach. But truly, such an inspiring achievement deserves a BLOG. Surely this will inspire many and may show a guideline for the ones like me who are struggling at a very low rating.

Once again — congrats to Shreyan Ray on this huge milestone!

This maybe off topic but i really want to know how are chinese coders that good. Among them jiangly can be considered as exceptional. I mean he is very much consistent in getting single digit ranks in recent contests and the number of Wrong Answers he gets on problems is very very low compared to others. How is he that good? I have watched many of his streams and the way he gets the main idea of solving complex problems so fast is truly fascinating. Do they practice problems from any chinese website to get that much good? If any one knows please do share. I could not find much details about jiangly anywhere. His mind is a mystery.

Hello people. I have encountered many expected number problems throughout this journey but i ignored them as they were too tough for me. but now i started learning about them. yesterday's atcoder contest problem F was a problem of this kind. but this was a little different compared to how the sushi problem of atcoder edu dp contest was solved. so there are a lot of variations and i am struggling a lot to solve them. so i wanted to ask you from where did you practice these expected numbers problems. Are there any places where i can get these problems to practice? Thanks in advance.

Hello coders
In most div2 contests i can't even do problem C if i have even 1 hour in hand. But many coders specially Grandmasters (not the LLM cheaters) and LGMs can solve them within 20 minutes or less. I have been doing cp for about 5 years now but still i am nothing more than a pupil. I know its practice but is that truly the only thing ? I am willing to give time to it but no matter which way i try to solve i always comeback to square one. I was struggling with Div2 C 2 years ago still today the same thing.. I solved many of them in my other account but if i chose 10 level C problems i can solve 3 of them i would say (that too with too much struggle) and the rest are just clueless for me. But i can do the first two problems in 30 mins to 1 hour but that's not good enough for me in comparison to the time i am giving to it. I want to solve at least 4 problems of Div2 in the whole time. I need to improve fast as I have given so much time to it. I wanna give this a final shot. Many people could even reach master or Grandmaster if they had put the time that i gave in cp. but somehow i am still a green coder. So can someone share your way of solving problems if you have overcome the phase where i am at right now ?

I am not bluffing. I meant everything I said here. You can clearly verify that by looking that this is not an AI generated post but written by a human. So Should i practice in other judges? Should I do advance problems of CSES after watching editorials ? or learn new algorithm ? please help me out. I am so pissed off at this moment.

Thanks a lot for your time in reading this. And sorry if i wasted your time but i had to post this as i dont know what to do next afterwards.

Fortunately I was able to solve all the problems of CSES sliding window section except two. One of them is this : sliding window advertisement.

I ran out of ideas to solve this. Any one who knows how to solve it please share your idea with me. Help me out. Do you think it can be solved if i have solved the other problems ? or do i need something different for this ? Thanks in advance.

Problem : Sliding Window Or

I am getting a TLE in this problem. This is my code:

// HELPPP...
#include<bits/stdc++.h>
using namespace std;
vector<long long> v, ors;
int main(){
 int n , k; cin >> n >> k;
 int x , a , b , c; cin >> x >> a >> b >> c;
 long long p = x;
 for(long long i  = 1; i <= n; ++i){
    v.push_back(p);
    p = ( a * p + b ) % c;
 }
 long long ori = 0;
 vector<int> bits(32);
 for(int i= 0 ; i < k; ++i){
  ori |= v[i];
  for(int j = 0 ; j < 32; ++j ){
   if(v[i]&(1LL<<j))bits[j]++;
  }
}
 ors.push_back(ori);
 for(int i = k; i < n; ++i){
  ori |= v[i];
  for(int j = 0 ; j < 32; ++j){
   if(v[i]&(1LL<<j))
      bits[j]++;
  }
  for(int j = 0 ; j < 32; ++j){
   if(v[i-k]&(1LL<<j)){
      bits[j]--;
      if(!bits[j]) ori -= (1LL<<j);
   }
  } 
  ors.push_back(ori);
 }
 long long ans = 0;
 for(auto i : ors)
     ans ^= i;
 cout << ans << endl;
}

How can i avoid the inner 32 loops in for loop ? or is there any other way to do it ?

I was trying to solve this problem from cses that is the maximum xor of a subarray of a given array. Can someone please help me to solve it. I tried for sometime but could not find any way.

This is the problem : Max Xor Subarray

Also it will be very helpful if you could give hints on solving the maximum Xor of subset of an array which i think is somewhat similar to the first problem. This one : Max Xor Subset

Thanks in advance.

iamnoob7 is a cheater. He cheated multiple times in the previous contests and got many skipped submissions. He cheated blindly in the yesterday's div3 round 1020 contest. Just look at his two submissions of C and you will see.

SpyrosAliv MikeMirzayanov

this is the problem — https://mirror.codeforces.com/contest/617/problem/E

I can't understand the solution. What is cnt[v] actually storing here and how are we getting the ans with simply cnt[v xor k] for one shifting of bounds in O(1) time ?

Thanks in advance.

According to you what should be the rating of someone to be able to solve all questions from div3 and from div4? I know it depends on question and various other factors but in general if you think what is your opinion on that?
You can share div3 and div4 opinion separately.

Hello There. I have seen many coders using Mint in their code. I have searched for it but could not find anything that useful. Can anyone explain what is this and why they use it so frequently? If anyone can share any link to it or something it will be helpful. Thank you.