This time is magical.

Surprisingly, it started at 12:10 noon on Sunday.

qp

This new start time is very good.

Hope this round won't be as bad as the last one...

pretty early for me, but i will be there!

all the best to everyone :)

Hi, everyone. I surprisingly discovered that yesterday was the 15th day of the seventh lunar month on the lunar calendar. Which is the Zhongyuan Festival in China or maybe the Chūgen Festival in Japan. It's the ghost fesival in China which means the gates of hell are opened up and ghosts are free to roam the earth where they seek food and entertainment. We go on the road and burn paper to guide these goasts So, I think it's the reason why they put off the contest until today. The time 11:10~13:50 (UTC+9) is also the peak of masculine energy according to some mystery.

(It there's anything incorect about the introduction of Chūgen Festival in Japan, welcome to leave your comment.)

So surprisingly that there is a similar contest on the same time. https://atcoder.jp/contests/jsc2025advance-final

Rlly hope I don't sell this time ;-;

I hope this game will be much better than last time

qp

qp

qp

Great round. Will the contest still be rated, even with so few participants?

Bro I cannot solve problem D.

Today's problems after C were better than yesterday's ARC imo. Especially; the problem F >>

I guess they have swapped ARC and ABC :))

Has anybody solved using divide and conquer for E?? If successful, please share it.

The solution of C is always surprisingly short.

Is there any particular way to solve problem E without using RNG?

why this giving error , this problems like c are so bad just one line code if someone guess 10 sec to solve if someone not no matter what he do always wrong i first utilise the b then simply take contri of a and c , any case you think about ??

void solve() {
    // executing code from here 
    int t;cin>>t;
    while (t--) {
        int a,b,c;cin>>a>>b>>c;
        int ans=0;
        ans=min({a,b,c});
        a-=ans;c-=ans;
        // dbg(a);dbg(b);
        if(a>c){
            // dbg(min(c,a/2));
            ans+=min(c,a/2);
        }else{
            ans+=min(c/2,a);
        }
        cout<<ans<<endl;
    }
}

Why problem F couldn't be solved with Dijkstra?

#include <bits/stdc++.h>

using namespace std;

constexpr long long INF = LLONG_MAX / 2;

int main() {
    ios::sync_with_stdio(false);
    cin.tie(nullptr);

    int n, m;
    cin >> n >> m;

    vector<int> W(n);
    for (int i = 0; i < n; i++) {
        cin >> W[i];
    }

    vector<vector<int>> G(n);
    for (int i = 0; i < m; i++) {
        int u, v;
        cin >> u >> v;
        --u, --v;
        G[u].push_back(v);
        G[v].push_back(u);
    }

    vector<array<long long, 2>> dis(n, {INF, INF});
    dis[0] = {0, 0};

    priority_queue<array<long long, 3>, vector<array<long long, 3>>, greater<array<long long, 3>>> pque;
    pque.push({0, 0, 0});

    while (not pque.empty()) {
        auto [d, w, u] = pque.top();
        pque.pop();

        long long nw = w + W[u];
        for (auto v : G[u]) {
            long long nd = d + nw;
            if ((nd < dis[v][0]) or (nd == dis[v][0] and nw < dis[v][1])) {
                dis[v] = {nd, nw};
                pque.push({nd, nw, v});
            }
        }
    }

    for (int i = 0; i < n; i++) {
        cout << dis[i][0] << "\n";
    }

    return 0;
}

Can anyone prove my solution for F? I claim that every vertex only have $$$m$$$ pairs of (current weight, fuel used) being useful , but I can't prove it.

my solution

all nice problem！！！ althought I just solve 3，but solve the rest problem with Editorial help me learn a lot！！：)

Is it fair to have random in E? The same solution can be judged differently just by random.

For problem D, first fill vector by $$$\frac{k}{2^n}$$$. Then keep incrementing numbers in even position by 1 alternating between left and right, continue till you have n%{2^k} remaining. If there is still some left then do the same for odd position. The imbalance will always be 1 when there is remainder other than 0.

Why this solution is incorrect??

problem D spj maybe wrong? Without considering K.

https://atcoder.jp/contests/abc422/submissions/69146833

There is a discuss of D's checker was wrong :Portal

For Problem E, there’s a minor issue in the editorial code: https://atcoder.jp/contests/abc422/editorial/13837

Hack:

3
0 0
1000000000 500000000
-1000000000 1000000000

If the algorithm randomly selects the last two points and computes the line coefficients, it gets a = 5e8, b = 2e9, c = -1.5e18, so |c| > 1e18, which yields a wrong answer. However, the same line has a valid reduced triple (1, 4, -3000000000) after dividing by gcd.