off-topic : why not make luogu an international platform just like atcoder ? Atcoder also was only for Japanese people at first , then it converted into international platform

My opinion: using AI should not disqualify the users. Why bother with it? Too much work with false positives. Lets just hope AI does not get good enough for Div1/Div2 :)

PinkieRabbit why you don't make it English??

I don't think codeforces does anything about AI generated code, because its rules state that generated code is allowed if:

the code is generated using tools that were written and published/distributed before the start of the round.

And ChatGPT is definitely counted as one of those tools.

It's not a problem.. in (div3,div4) most people and ai models can solve a,b,c but only a real person can solve d,e,f in my opinion the problem setters can make a small trick in the problems makes the ai models not able to solve'em

Make problems difficult enough or contain corner cases(or smth else) so that ChatGPT won't get AC.

If chatgpt can solve the problem correctly then even a monkey can.

Can a language model do an anti-plagiarism check?

PinkieRabbit would you share which algorithm you implement in anti-plagiarism module. I googled but could not find a promising solution. We can't use jaccard, cosine similarity as they do not catch sequences leading to more false positives. Could only think of LCS on tokenized code submissions. How do you deal with false positives? There are thousands of submissions and there is high chance two user thinking same sequences.

I've been contemplating sharing my observations on a this topic, and I believe it's the time to share things anyway.

I have conducted several experiments using LLms to assess their performance in various programming contests. Among the models I tested, GPT-4 has consistently stood out, hence I predominantly utilized it. To enhance its capabilities, I integrated an agent on top of GPT-4 that automates the processes of code writing, testing, and evaluation. Additionally, I employed prompt engineering to simulate a chain of thought.

The results have been quite astounding. I allowed the model to participate in three contests autonomously, and it achieved impressive scores of +1900 and +1600 in div3 and div4 contests "fully solved a Div4", respectively. I've also observed that the manner in which the problem statement is presented to the model significantly impacts its performance. On occasion, minor interventions are required to guide the model to the correct solution.

For those interested in replicating my findings, I recommend using ChatGPT Plus combined with a code interpreter. The outcomes should be commendable. As a side note, my current configuration of the model has a rating of +2000 on LeetCode and even achieved a +2400 performance rating in one of the rounds.

I'm sharing this to highlight the potential of such models. While many believe these models don't make a significant difference, I speculate that an advanced model beyond GPT-4 could potentially solve even Div1 problems.

PinkieRabbit there is an youtuber who uploads solution in youtuber shorts at contest time. There are many cheaters who took help from it. Can't we stop them?

I guess use AI modal to check the submission of a particular contestant like how he/she use the write code in there previous submission including all the previous contest he/she had given before.

chatgpt can solve some problems that most people can solve it but I think in the future it will be able to solve difficult problems that need advanced techniques so we must solve these issues before this happens

My opinion is that using language models to generate code is not cheating. They are legitimate tools for code generation, even if their performance is lacking at the moment. All they do is lower the "skill floor" of programming.

If you think about it, higher level languages like python already do that: you can write powerful code without thinking about memory allocation or integer overflow or ...

They can be used just as well in contests as in real-world problems outside contests.

Thanks PinkieRabbit for taking the steps to prevent cheating, truly amazing!

I just used ChatGPT to help solve the uva problem 1600.

https://onlinejudge.org/index.php?option=com_onlinejudge&Itemid=8&page=show_problem&category=0&problem=4475&mosmsg=Submission+received+with+ID+29688775

Prompt:

        Pos p1 = new Pos(0, 0, 0, k);
        ss[0][0].add(new State(0, k));

        PriorityQueue<Pos> pq = new PriorityQueue<>();
        pq.add(p1);
        int ans = -1;
        while (!pq.isEmpty()) {
            Pos p = pq.poll();
            if (p.x == m - 1 && p.y == n - 1) {
                ans = p.dist;
                break;
            } else {
                for (int d = 0; d < 4; d++) {
                    int ndx = p.x + dx[d];
                    int ndy = p.y + dy[d];
                    if (ndx >= 0 && ndx < m && ndy >= 0 && ndy < n) {
                        if (ss[ndx][ndy].dist > p.dist + 1 || ss[ndx][ndy].k)

                            if (grid[ndx][ndy] == 0) {
                                if (ss[ndx][ndy].dist > p.dist + 1 || ss[ndx][ndy].k < k) {

                                }

                                ss[ndx][ndy] = p.dist + 1;
                                Pos np = new Pos(ndx, ndy, p.dist + 1, k);
                                pq.add(np);
                            } else if (p.k > 0) {
                                ss[ndx][ndy] = p.dist + 1;
                                Pos np = new Pos(ndx, ndy, p.dist + 1, p.k - 1);
                                pq.add(np);
                            }
                    }
                }
            }
        }


---

1600 Patrol Robot
A robot has to patrol around a rectangular area which is in a form of m × n grid (m rows and n
columns). The rows are labeled from 1 to m. The columns are labeled from 1 to n. A cell (i, j) denotes
the cell in row i and column j in the grid. At each step, the robot can only move from one cell to an
adjacent cell, i.e. from (x, y) to (x + 1, y), (x, y + 1), (x − 1, y) or (x, y − 1). Some of the cells in the
grid contain obstacles. In order to move to a cell containing obstacle, the robot has to switch to turbo
mode. Therefore, the robot cannot move continuously to more than k cells containing obstacles.
Your task is to write a program to find the shortest path (with the minimum number of cells) from
cell (1, 1) to cell (m, n). It is assumed that both these cells do not contain obstacles.
Input
The input consists of several data sets. The first line of the input file contains the number of data sets
which is a positive integer and is not bigger than 20. The following lines describe the data sets.
For each data set, the first line contains two positive integer numbers m and n separated by space
(1 ≤ m, n ≤ 20). The second line contains an integer number k (0 ≤ k ≤ 20). The i-th line of the next
m lines contains n integer aij separated by space (i = 1, 2, . . . , m; j = 1, 2, . . . , n). The value of aij is
‘1’ if there is an obstacle on the cell (i, j), and is ‘0’ otherwise.
Output
For each data set, if there exists a way for the robot to reach the cell (m, n), write in one line the integer
number s, which is the number of moves the robot has to make; ‘-1’ otherw

---

help me finish code

ChatGPT:

      PriorityQueue<Pos> pq = new PriorityQueue<>();
        pq.add(p1);
        int ans = -1;
        while (!pq.isEmpty()) {
            Pos p = pq.poll();
            if (p.x == m - 1 && p.y == n - 1) {
                ans = p.dist;
                break;
            } else {
                for (int d = 0; d < 4; d++) {
                    int ndx = p.x + dx[d];
                    int ndy = p.y + dy[d];
                    if (ndx >= 0 && ndx < m && ndy >= 0 && ndy < n) {
                        int newDist = p.dist + 1;
                        int newK = grid[ndx][ndy] == 1 ? p.k - 1 : p.k;

                        if (newK >= 0) {
                            boolean shouldUpdate = true;
                            for (State state : ss[ndx][ndy]) {
                                if (state.dist <= newDist && state.k >= newK) {
                                    shouldUpdate = false;
                                    break;
                                }
                            }

                            if (shouldUpdate) {
                                ss[ndx][ndy].add(new State(newDist, newK));
                                pq.add(new Pos(ndx, ndy, newDist, newK));
                            }
                        }
                    }
                }
            }
        }

but case 2 is wrong.

Prompt:

4 6
1
0 1 1 0 0 0
0 0 1 0 1 1
0 1 1 1 1 0
0 1 1 1 0 0

this input should output 10 , however -1 

Prompt again,

int newK = grid[ndx][ndy] == 1 ? p.k - 1 : p.k;

when grid here = 0 , it should be k .

Then it gives correct code to pass.

In summary, I tried the solutions for several hours. And then used ChatGPT for some fixes. After some prompts, it gives the correct code to pass.

Full conveseration:

https://chatgpt.com/share/f24e2525-0318-4c46-8f77-743d664330c0