So yeah, hey guys. I'm Dragmon, the "cyan author", the "trash problemsetter", the "author of Codeforces Round 1033 (Div. 2) and CodeNite 2025". So, you might be wondering, who am I to suggest anything to you? Well, I don't know, you clicked on the blog. I have made sections, you can just skip to what you want to read.
Brief History of Me
First of all, I'll give a brief history of myself, because why not? You might not be expecting it, but problem-setting is my passion. I've been setting problems on Physics and Maths during my JEE time. If you are interested, you can find them here in drive. I had ~100 more in brilliant.org, but sadly, the community feature was disabled around July 2021. If you are interested, I can get it exported and share it with you. After coming to IIT Kharagpur, I joined CodeClub and set some problems in the unrated CodeNite that we conduct half yearly, along with amazing peers, some hard, some easy, you can find them here in doc if you are interested.
How can you set a problem?
So, why can an expert(lol no, specialist) like me set problems that are 2600+, you might be wondering? I think it is because of one simple fact: to be a 2600+, you have to master most topics and problems that are 2600+. But to be a problem setter, you can master just some/one topic, and it will work out for you. For me, the topic turned out to be "Graph Theory". I've been fascinated by this topic since I set foot in it, maybe during some Div 1 codechef long challenge in 2019, when the first two problems had graphs/trees, so I basically had no option. Since then, I've really taken it and I do have a publication on the topic of "Line Graphs" link to Springer paper if you are interested. So, hard questions that I come up with typically come from the domain of graph theory, especially line graph, like the problem 2120G - Eulerian Line Graph. I would like to thank the problem setter of problem "LLLGRAPH" gainullinildar, a.k.a. 300iq for the ‘April long challenge 2020 in CodeChef for getting me inspired and pursuing the topic of line graphs(Yeah, I'm thanking him here because I never got another chance).
The thing that I do to come up with problems is just find problems in everything in life — while learning topics, solving problems(general version of problem, combining it with some other concept, array problems now on trees, graphs), etc.. For example, I'm an ardent gamer. I love playing games, especially ones that involve RTS, RPG, or Supercell's name ;). I've come up with problems with their mechanics and stuff. Also, real life is just as good for that stuff of finding problems. One of my problems is even based on screen pattern lock(you can find it in the doc above).
As an example, for the contest that you might’ve not liked Codeforces Round 1033 (Div. 2) and CodeNite 2025, I was the author and setter of 2120A - Square of Rectangles, 2120D - Matrix game, 2120E - Lanes of Cars, 2120F - Superb Graphs, and 2120G - Eulerian Line Graph. 2120A - Square of Rectangles I just came up with randomly, 2120D - Matrix game was based on an interview problem that I got asked about whether we can find a rectangle in every infinitely colored grid, 2120E - Lanes of Cars when my friend was telling me his problem, but I somehow misheard it, 2120F - Superb Graphs when I was studying the topic of conflict-free coloring, and came across the concept of neighborhood diversity and type graphs, 2120G - Eulerian Line Graph when I was studying and research line graphs.
Defend for the round
Now, if you are thinking "well, where is the defend for the round Codeforces Round 1033 (Div. 2) and CodeNite 2025?". Well, as is said in One Piece, you can't hate the ship you made, even if the whole world is against it, so I love my problems and would continue to. Also, as I like to say, trying to change a person’s opinions just leads to a void function, that can't return anything to you or the said person. So, please be your own judge. An exact match for the problem by my fellow co-author harshith_04 being found on the last day leading to the round being a div. 2 instead of div.1 + div. 2, when searched by our coordinator, was really unfortunate, and I won’t blame anyone for it. Believe us, we were wayy more hurt by that news than anyone of you even think you were. (Though it seems that you’ll would’ve hated it even more if it was a div.1+div.2 from the comments, blog and messages you’ve sent me, which yes I’ve read them all. It’s funny how people can be damned to hell for the "heinous crime" of setting a “trash” round and being an Indian.)
Issue of cheating
As for the issue of cheating, I just completely agree with Um_nik's blog. I don't care if people cheat or use AI. I came to this platform to enjoy solving problems with a time constraint, not to boast my rating or show it to a company. And I'm not setting problems that I don't like just to satisfy the criteria of it not being able to be solved by AI (Some people are saying that we authors leaked the problems ourselves lol. Do people really think we’re that stupid to sell off our own problems?).
And to the people who are accusing my co-author harshith_04, I just don’t get what a person cheating or not cheating says about the quality of problems they prepare, least of all one that happened in 2023. For all I care, he is just an amazing, cheerful, and helpful senior who loves Competitive Programming. (Maybe you don’t respect me now, aye? Feel free not to lol, I ain’t going as shallow to just judge a person by one incident, which I’m not sure is true and neither do I care if it is).
Conclusion
As for this blog, feel free to like it or dislike it, idk, not like my contribution is gonna get affected much after my announcement blog for Codeforces Round 1033 (Div. 2) and CodeNite 2025. But, in any case, I'm happy if this blog helped maybe even one of the experts/cyan coders are to set problems, I'll also like to thank all the problemsetters as it is only because of them that the platform is still running. Don't worry, I don't have much plans for rated problemsetting in the future :). I think I have my job done! Thank you for participating in this with me, I guess :)) (Did you even read the whole blog lol?).
Auto comment: topic has been updated by Dragmon (previous revision, new revision, compare).
No matter what people say, I personally loved the problems. They were very good. I didn't know about trees so I assumed that I won't be able to solve Problem C but the problem didn't require any explicit deep knowledge of trees. I came back a few minutes later and started learning the structure of trees in between the contest and in the end I did make a submission on C but it was wrong. Still, it was fun to make the transition from thinking I'd not be able to solve C to trying to code it
you both got trashed just because you shared the nationality with a bunch of lowlife folks; every round has problems that are disliked by a set of people so imho filter the feedbacks that you received
i wouldn't say it was a very good round, but decent problems, and it surely didn't deserve the criticism thrown at y'all!
I appreciate this blog, and I thank you and the entire team's time for preparing these problems as what happened already cannot be changed.
My question is due to the excessive cheating in this contest with ChatGPT, are there any actions you as an author of the contest plan to take in order to skip/ban all the cheaters?
He did not take any effective action. He could have sanctioned the cheaters during the competition time. Even SSerxhs (schzmooth coordination) asked during the match whether he should be the one to ban the cheaters. It could have been handled perfectly, but the fact is he took no action at all.
We were not granted permission by KAN to sanction the cheaters during the competition time, so we couldn't have done anything.
First, I apologize for my words—I thought you could take action but didn’t, when in reality, you couldn’t. I recall that during coordination, Little_Sheep_Yawn or others mentioned that some problem implementations might be very short and easily exploited by AI. Since you knew this could lead to widespread cheating, couldn’t you have proactively requested permission to ban cheaters in advance?
This is the first time I am problemsetting a rated round, and I didn't have any idea that authors are allowed to ban cheaters. I only got to know about it during the round when one of the testers told us about it, but as I said before, we weren't granted the permission.
No worries, so am I correct that the only remaining option is just to wait for plagiarism checks?
I like how you know harshith_04 is a cheater, but you still stay diplomatic to not ruin your relations with your senior. Also the issue is not a specialist framing problems, the issue is cheating, if the contest was leaked before hand.
For the most part, coordinators and authors do not care about whether approved problems are gptable or not.
To quote from the blog:
Ohkk, thats quite reasonable.What about the fact that blog quotes, author is here to enjoy cp and not boast his ratings, Then why does he allow someone who is cheater, who is here to boast his ratings and make cp less enjoyable, be his co-author? And why would anyone trust such cheater, that he might not have leaked the problems to fellow cheaters who too wanted to boost their ratings? Cause I know such dudes all they care about is getting a rating at which they could get shortlisted in some MnCs screenings.
First of all, as I said in the blog, I still am not confirmed if he cheated, neither do I care. I know him way too much personally just to think that he will sell off the problems, and he reached cm on his own, so I don't know what exactly is the matter. If you really think he leaked problems of the contest he took the initiative to host and put all of his efforts into for 1 year and no less. I don't think I can change your opinions, and I don't think I want to, he is smart enough and doesn't need approval of anyone I feel.
He has two skipped contests, and we don't know how many did go un-noticed. I've got no problem with your part i.e. of problem setting, except letting a person with not so good history be a coauthor. I know you have attachments with the guy irl, so I dont even want you to publically accept that he's cheat and trusting such a cheat might've turned into a bad idea.
You may have truly been deceived by him. Problem 2104F was a 2600-rated question, and the DE problem was an extremely difficult DP challenge—yet in 2113’s Problem C, he somehow got stuck and couldn’t solve it. Problem C in 2113 was something even I could solve; it only required knowing prefix sums, and most people would crack it within five minutes. In my community, I had others test this problem, and it turned out to be exactly the kind of question AI would misunderstand and fail to solve. This is undeniable suspicion. I think... he might really have leaked/sold the problem... Though I’m more inclined to believe you were simply too trusting, which made you easy to deceive.
It is not unusual for people to perform badly in contest while being a strong problem solver outside of contests. I have seen such examples in my college, people who are barely pupil (mostly newbie) but can solve problems till expert rating (from wide categories, not just some topics). It is not unreasonable to think his co author may be that sort of a guy.
Also, aren't there other authors who are low rated (expert or something) but create higher rated contests? That in itself is no ground to accuse anyone of cheating.
You're wrong. What I mean is that precisely because problem 2113C couldn't be solved by an LLM, he can't solve it. Meanwhile, in the 2104 contest, the LLM could solve more problems. If your combinatorics were slightly better, you'd probably understand what I'm getting at.
That's a good point even though I personally don't consider it incriminating enough at all. Could be luck.
I don't know why you are downvoted so much though, that ending was a good roast :D
Maybe what I said was a bit too harsh, and it triggered resentment from a certain group of people (but I had to say it). The probability of all these things happening together — P(that he has a prior cheating suspicion) × P(that he performs well in rounds where the suspected LLM also performs well) × P(that he performs poorly in rounds where the suspected LLM also performs poorly) — and on top of that, the probability that he didn’t actually cheat, I guess it’s less than 1%. You’re right that this alone doesn’t constitute sufficient evidence. Maybe it’s just luck. But I really can’t convince myself otherwise; it feels like the tiny chance of a DNA test not confirming kinship because of a hash collision. (Or maybe it really is? Haha.)
This is not targeting anyone or racial discrimination. I only have disdain for people with prior cheating records. They must repent for their past unethical behavior, because they are the ones who ruined the environment of algorithm competitions.
I agree with your formula, I don't agree with the values. When I said it can be coincidence, I meant that what you said "he performs well in LLM contests and not in others" is biased. I haven't tracked him through all contests, and assuming you have also tracked him for only 2-3 contests, the result that "he performs well in LLM contests" maybe false / not as strong as you are implying.
If he really does perform well in every LLM contest and not in others, then indeed it is a strong case.
As you said, I’m unable to find earlier versions before the LLM training for testing, so there is no way to verify this.
hii cry
They do, if you've seen past rounds. This blog, for example.
Even our got gpt tested by one of the fellow testers, as you know, and problems from D and beyond weren't solvable by GPT o1.
meanwhile o4 is launched.
I'm sorry, I read the tester's feedback who tested it, he used o4-mini, not o1.
Problems weren't rejected just because GPT could solve them. It just so happens GPT happened to do badly on that round, probably because of chromate00's unusual problemsetting style lol
Well, I was sure I had some hard evidence to back up my claim, but now I can't find it. However, there still appears to be anecdotal evidence.
that was back when we thought AI still mega sucked and there was no way it can solve past B
First of all, it was a special case (1000th round or something) as the author says in the blog (chromate00), and secondly even in that contests people where able to cheat (you can read the comments)
Herd mentality runs deep, even among the coders; no one seemed willing to think how the problem setters could be at fault for the widespread use of AI agents. Ironically, directing their frustration at Sam Altman might have made more sense. Here's hoping this doesn't dampen your spirits. This platform still needs new contests, for which it needs problem setters.
Maybe one day there will be another round under your name, and hopefully people will realise you don't have to be red to create problems that can't be solved by reds
- people will realise you don't have to be red to create problems that can't be solved by reds
Do you honestly think that's even possible?
Yes. There are many instances.
Yes obviously it is, Red is like CEO of a company, he need to be able to manage 100 things at once, while the problem setter is like head of some department, he just has to focus on one domain, and he just needs to be good in domain he is.
May be
Off the top of my head, 2035H - Peak Productivity Forces was written by a CM, and if I remember correctly, which only had 1 official solve in the contest. It's also very much possible for someone to come up with an idea, struggle to solve it, then hand it off to someone much higher rated who is able to solve it.
nice, good to know,
You are right. The issue of the contest wasn't about the problems,it is due to some Pupils ans unrated getting higher ranks than Expert and GM.
I personally know Dragmon very well and he never sets problem for any recognition.
I will say downvoting the round because of cheaters going out of control is stupid — do people not realize that authors cannot control who cheats on their rounds?
However, some of the problems were really not up to quality. I personally got frustrated with problem E (you can see my submission details to understand why), but I simply don't think A, B, and E is acceptable in any div2 round (much less a combined round that this round was going to be). C and D were debatable — C I actually found quite nice, D I personally didn't like but I know there are lots of people who did like it. I have not read F or G so I cannot comment on that.
Maybe some time should have been spent trying to come up with better problems. In an ideal world the announcement shouldn't have >2000 downvotes, but maybe even out at around [-100,100] range.
E was a problem that had mixed opinions during our testing, and one extreme sides. Some people like it quite for the details in implementation it needed and to solve exactly, and some hated it for the same reason. About A and B, we received pretty positive opinions about it during testing.
About the downvotes, they started pouring when the round became a div. 2 from the intended div.1+div.2, and only worsened when the whole cheating scandal happened.
Please don't think I don't respect your opinion, I totally do! I'm just explaining how the problems made it to the contest.
what are your opinions on cheating scandal? what do you think, how it happened?
I was able to come up with a solution for A,B,C and I think they were quite good, problem A was reasoning based (I saw someone hating it for l3<=l2<=l1, which is funny) and someone hated B because the limit was not matching with the complexity of the solution (funny again). The massive hate the round got isn't because the problems were bad or due to the junky leaderboard, Someone even wrote a hateful blog before the contest , and rest was the job of distrust.
yeah, i dont know why are people hating the author for the cheating that has happened.
.
Although I also agree with Um_nik's blog, it is rather disturbing and frustrating to witness numerous cheaters seizing the scoreboard. While I know it is hard to do plag check and that is why we have rating rollback, it is abnormal to let these obvious cheaters stay in the standing and be removed only after rollback that occurs several weeks or months later.
you are not getting hate and this accusation because some cheaters solved it before some LGMs. you are getting this hate because you are INDIAN. There are cheaters in every round. even saw pupil and specialist people getting global rank 1 in recent rounds but no one said a word but the moment any Indian is in problem setting the Mindeveloped guy started his agenda lmao
Funny how you can put fuck india in your profile and CF wont take any action but the moment dominater used RETARDED word in his comment everyone baffled
It's quite disappointing how some groups form communities around certain agendas and then portray themselves as victims whenever they face any criticism or backlash.
if you ever look around the contest rating and search cheaters you'll see that there are more number of chinese cheaters in top 100-500 in recent few contests. they cheat on this platform called bibli i dont know the exact name but only indians face the backlash. why? because a retard RED tag user is racist all the time and is spreading hate.
yesterday the problems might be gpt-solvable i dont know i didnt participate and actually in every contest there are gpt solvable but yeah he saw the oppportunity to farm ++ contri and being racist
Bilibili is a video website in China (like youtube). I don't think anyone can cheat on it.
Cheating happens on youtube too. I have seen live videos with telegram channel links or website links that sell solutions, so your claim that cheating can't happen through video websites is bs.
But on bilibili I never heard of such things. If you don't trust me, you can try to search here by yourself. Saying "there are more number of chinese cheaters in top 100-500 in recent few contests" is also ridiculous.
As for "they cheat on this platform called bibli", that is totally nonsense. If we really want to do so, we have better ways, instead of publishing the groups publicly like a fool.
In fact, the reason why so many people in China have red names is that at various stages—middle school, high school, and university—they can secure direct admission opportunities based on their rankings in algorithmic competitions. In reality, cheating in China is very easy to report, and even a slight violation of the participation rules can result in losing the chance for direct admission. Because cheating is virtually impossible and the academic environment in middle and high schools is extremely stressful, many turn to algorithmic competitions... This is why only in China does the competition become so intensely obsessive
Chinese folks do take everything to the extreme lol, your IPhO teams are always crazy. There is a similar obsession to the Engineering Entrance Examination in India. However, our schools are not so good and thus, coaching institutes have monopolized the JEE into a rat race.
Both are wrong, specially the fck india guy should be banned, what he is doing is not revolt against cheaters, it's just pure racism.
If you really studied the standings of past contests, you would find that most of them are alts of high-rating users. I wasn't saying having alts is right, but that's different from this one.
Cheater number in the first page this contest is almost the largest one among all contest. I felt codeforces was losing control for the first time. I don't think the point is simply that, the writer is an Indian (although it may lead to some stereotypes).
Many other contest has Indian writers, but we didn't doubt the writers. So I don't consider the anger of us was only because of the fact that writers are from India. I am also against racism. We are angry because of the unusual standings itself, not the nationality.
ofcourse. when an adhoc round is full of cheaters, nobody remotely cares. but now a round filled with math problems is full of cheaters, suddenly everyone are on their nerves. wtf. hasent this cheaters issue been a concerning thing since idk last 6 months or so.
problems were enjoyable. now you couldnt solve it is making you salty. people saying, pupils and specialist are above GMs, no shit sherlock. it is LLMs at the back. and you know, that LLMs can do math and implementation problems easily. So, i'll just say, its a skill issue. the justification for this skill issue is: the same treatment experts got 6 months back when LLMs started to solve upto them, they were responded back with a plain and simple "skill issue" and i kind of agree with that. we are here to improve ourselves. but now, the same thing is happening at higher levels, they are getting salty.
about the problems quality, no one talks about a full ad-hoc round's quality. but when theres suddenly math, we are getting some interesting reactions.
again, SOME people (even higher rated users) need to chill tf down. this was same as a regular cf round taken A BIT extreme by LLMs. this has been an issue since last 6 months. no need to make this a bigger issue.
nobody is complaining about the math problems they are complaning about impl hell E and braindead A and B. also the LLM cheating was a lot more extreme this round, so ppl are suspecting problems were leaked beforehand.
LLM cheating will be there. you know that implementation problems can be easily done by LLMs. I agree that A and B were easy, but there are some contests where A and B are braindead too, right? then why shit on this round? i dont think that the authors would leak problems. they know that if they leak, there will be a huge amount of backlash. so, to sum it up, LLMs doing the implementation and maths is the only issue here. if there is something i'm overlooking on, do tell me.
As I said, that's not only about LLMs. Half of F codes on the first page begin like below:
And the whole
TwoSATis almost the same.ohh, didnt know about this. sorry, i missed this. then it is a concerning issue. my bad.
could be the case of telegram channels people are referring to.
Do you know the reason, why ratings have not been updated yet, For div 2 they are usually updated 2-3 hours after contest end?
it became unrated
Edit: Sry ratings are updated
Please don't be discouraged :( I know prepping a contest under pressure has a lot of work, and finding out that there is a coincidence in your problemset last minute really hurts (I experienced this for myself in Codeforces Round 965 (Div. 2)).
Please continue the positive mentality and I look forward to seeing more stuff from you in the future.
It's considered unacceptable to use words like 'fing retard,' yet it's somehow acceptable for people to create groups with names like 'F India.' It seems as though they believe it's their inherent right to always be favored, regardless of who's actually at fault.
I don't think your problems received "positive" feedback from the coordinators in terms of "quality" either.
Quoting a recent comment by Dominater069, "My feedback on most problems were bad even for div2, and I flat out told this is not fit for div1"
Did you just think going ahead with the problems made sense, even after knowing the experienced setters didn't like them either?
I would share the feedbacks we received from testers if I could, but I don't think I can. We did receive negative, but many were positive. Some problems were removed because of negative opinions, and that's totally legit. About the part "Did you think going ahead with the problems made sense....", we as authors propose problems, deciding to go ahead with the problems is what coordinators are for. If you feel you didn't like the problems, I won't change it because that's totally legit, but I've also received positive feedbacks about the problems after the contest ended. So, all and all, I don't feel they were bad, but as I wrote in the blog, I can't hate the problems I've made.
what you are talking about is a completely different topic from what i'm tryna say i am not saying anything about question quality
How the fuck contesters are able to solve E under 3 minutes? While even gm and lgm are solving longer. Of course everyone thinking that someone in the contest-creating team has leaked problems beforehand.
E is free you just need to implement it but the implementation is super long and annoying, so LLM cheaters can solve it quickly while humans take a lot longer
you didn't write anything about the scoreboard. however i think the most important reason that the announcement being downvoted is the weird scoreboard. So now I still want to know what happened with this colorful Indian participants on the top of the board. I don't believe that these ones have evolved overnight.
another problem is that why you decide to cancel the div.1 part because I think there may be some other solutions just like play a 7 problem 1+2 or delay the round to make a new G.. cancelling a div.1 round is disappointing
third I want to ask my own question that WHY EVERYONE OTHER THAN ME(EVEN CYAN ONES) CAN MAKE LOTS OF PROBLEMS AND HOLD CONTEST BUT I CANNOT COME UP WITH ANY INTERESTING IDEAS???????
last one is the real question here ;_;
When are we going to get the Editorial?
To be honest, I genuinely enjoyed the problems in the contest. I was close to solving Problem D, which made the experience even more engaging. However, I was quite disappointed to see the Div. 1 + 2 round being changed to a Div. 2 only round. With the prevalence of cheating, the contest already faced serious integrity concerns, and this decision only added to the frustration. That being said, I believe it's unfair to judge or hate an individual based on assumptions about their integrity, especially when it concerns a contest they authored. Yes, the problems leaned more towards mathematical thinking, which some participants might enjoy while others may not. But accusing a problem setter of leaking problems without concrete evidence is unjust. It’s valid to question whether less experienced or lower-rated participants should be allowed to author Div. 1 + 2 contests. However, that discussion should not become grounds for baseless allegations. Unfortunately, cheating is widespread—it's said that nearly 90% of Indian participants engage in dishonest practices. With the recent advancements in AI, language models like GPT-4o and Gemini 2.5 Pro are capable of solving even hard problems. Still, the misuse of these tools by cheaters should not be used to tarnish the reputation of contest authors. Accusations should be based on facts, not assumptions.
Tumhe to fasi hogi...fasi
I enjoyed the problems, different style
I do not think it is fair that some authors simply deny responsibility with the notion on that, "I set problems that I like, I do not care about AI". In the end, you're setting problems for people on this website to solve, and most people here are affected by the fact that there are cheaters. Regardless, the amount of hate that you're getting is not justified at all.
no problem A cannot be solved with AI, and AI can basically solve A-D on most contests and usually E as well. but I do agree adding conceptually trivial problems with cancer implementation (which AI is the best at) should be avoided.
is the round unrated ? cause rating still not changed..
Yes, let me allow my cheating scum friend to set problems for a possible Div1 + Div2 round. What could possibly be wrong with that? Surely the problems will be of high-quality and non-ChatGPTable right? Right?
I dont know the authors personally, so i may be wrong. But as he said it happened in 2023, who knows he might have learned his lesson by now, we should atleast give them a chance to be normal again.
Just because he cheated long back doesnt mean we should keep these acusations forever right
bros personality fits his name
I don't think the author harshith_04 has given any clarifications regarding his cheating or skipped contests (at least not as of now)
Surely my friend won't leak the problems to his "IIT/Telegram group" friends so that everyone rating go stonks and no one will notice hehe. What's that you say? People <1600 rated are doing better than LGM's? It's just skill issue lmao git gud bro.
Is this round unrated now ?
A low rated user can create good problems, I agree.
However a cheater must never be allowed to become problem setter. A cheater uses shortcuts to boost his real life image, and writing "I set a hardest problem in Codeforces div 1" in their CV / linkedin will just feel too good for them to ignore the chance. And cheating in problem setting is so easy, just copy some old problems in some random online judge.
as expected.
Cheating in competition and cheating in problemsetting are too weakly correlated, so one does not become an immediate implication for the other. I can assert this as I have seen multiple (ex-)cheaters that did not cheat in their problemsetting career (and also the contrary — non-cheaters who plagiarized tasks and got blacklisted).
It's just your anecdotal experience against my anecdotal experience. Mine says a cheater in one context will usually cheat in other context, especially similar ones.
It's probably your feelings instead of anecdotal experience. On the contrary, I can give examples of actual cases:
As you can see, we have clearly enough examples where an author belongs to one category but not another.
Also the grimoire of code contest you've mentioned, that too is a club in IIT KGP similar to CodeClub which organized this controversial contest. The thing I'm trying to state is that I've seen many IITians trying to cheat their way into such things to continue the excellent academic past they have, I am also student of same college but still if I'm against my own of my own alumni, then there's some reason for it, such people also damage the reputation of my institute.
For this case I am completely confident that the problemsetters were honest and innocent, as opposed to Round 819. Can't say about problem quality, but at least they are not dishonest.
Without knowing anything about me, you decide to disregard my opinion on basis of my feelings (which may be true because some cheaters affected me personally). While I stopped participating in CF contests long time ago and rarely interact with CF, the last 10+ years I'm still involved in local communities by setting contests and other activities. I saw many people who cheated in online / offline contests, online judges, problem setting, teaching online courses, etc. And my personal anecdotal is that they usually cheat in more than one settings.
Your links won't change my opinion at all since it's usually impossible to catch smart cheaters — so just say an author who cheat in problem setting didn't cheat in contests doesn't mean anything — they can be smart enough to never been caught.
Thanks for the discussion and have a nice day.
See your point is correct, but if someone has a history of cheating, don't you think its normal to suspect them in future in case something they're involved in goes wrong?
If something goes wrong, admins can look into the matter independently. Does not necessarily have to do with previous cheating record in competition.
When a crime happens, the police first looks up the first suspects are the ones related to crime and with past history of wrong doings, it applies to each and every domain. Like I said and have been saying "suspect", so even if you might be innocent a bad past will always have you in the wrong spotlight.
He shouldn't be suspected unless you really have any proof. Do you have any personal grudge against him? Your only "proof" against him is that he cheated (in contest, no one knows if he ever cheated in problem setting) 2 years ago. Really? People change in 2 years, especially people in college.
Meanwhile there are so many other aspects regarding this which include LLMs, participants cheating etc., which have way more substance than your 2-year-ago-did-something proof, and you are just trying to make the author look bad. Get a spine, don't be jealous if that's what you are.
See he is my senior and I was very happy initially when the round was announced, you can check my previous comments. But the thing is,He couldnt solve a question in a recent question which was not solvable by gpt, some other user pointed all that out, also earlier everyone was denying that he even cheated in first, now you all say he cheated 2 years ago. In 2022 in a similar contest organized, by one of the other clubs on my college, one the question was plagiarized, So when this contest had sudden surge of cheaters there are could be only two reasons, problem leaks or very gptable problem, and seeing the history, I think it's normal to consider both cases.
I think 2022 situation is not comparable to 2025, because since then the cheating issue has been raised significantly and people have started to understand it's seriousness. Moreover, CF has always been more respectable than other platforms, so even cheaters (while problemsetting) would think twice before doing their mischief here.
Also, a strong point raised by many is, why would he leak the problems to stupid greens and cyans and blues, knowing for sure he would get exposed, when he had this one chance to fame? That doesn't make sense in any universe and you all keep conveniently ignoring this.
Brother I am not ignoring any reasons, the reasons are valid, all I am saying is people will obviously suspect people with a bad past. And no people have not started to understand it's seriousness, you too know how many clg students cheat to boost their ratings before placement.
I know how many college students cheat, you are from IIT, I am from from a tier 2/3 college, so I know better than you.
When I say "seriousness" of cheating I am referring to students who have reached the level that they can set contests on CF and publish algorithmic papers (that one of the authors did). Do not compare random cheaters to the problem setters in this case, I am sure they do not belong in the same category.
And lastly, you keep saying "people will say this" my brother is your name people? Are you society? Where is your opinion? Your brain? Do you speak for the society or do you speak for yourself? If everyone started thinking the way they think society thinks, their would be no progress. Or maybe that's how it already is, idk.
See when I say people, I'm referring to myself from your pov, I'm not related to you in away so I'm a random person, therefore I used the term people. Also The guy you're saying has published paper and the person accused of cheating are different, I never said that aryan cheated.
Then refer to yourself, don't hide behind people.
The guy is different but they are a team, the point is if they understand the seriousness of cheating. If they are this good friends, they do understand it's seriousness, that's it. They do not belong to your group of "college cheaters".
It's not hiding, my point is trying to tell you have the open mind to listen what "people" have to say, this is not an eutopian world where everyone will have same opinion as you. Also what do you want, if someone is a cheater and is with a good guy, he can't even be suspected? Cmon!
I think I made my point pretty clear on when someone can be suspected.
Also I have a pretty open mind, thanks for the concern :D
Also I have said the above things because you have tried to push your narrative at multiple places since the contest has happened. Either get some proof, or shut up.
Naah, before the contest started I wrote many positive comments too, ill obviously comment what I feel like, obviously won't fabricate my thoughts.
Also to me someone who has cheated in the past will always be a cheater. Anyone whose cheated in past, whether in competitions, exams, relationships, anything, they will always be considered a cheater, if you don't think so, then it's your pov I don't mind it.
Sure, keep the opinion. I am not trying to change it, I am only saying you should keep your opinion to yourself because it is a stupid opinion.
I have cheated in school test and exams, but of course I haven't cheated in my JEE, interviews, CF, or anywhere else. Cheating once or one place does not imply other, unrelated things.
(Never got the opportunity to cheat in a relationship because I have been in none, or would have added that I never did that either too TT)
Naah, the contest was public, cheaters who cheated, cheated in public, the platform is public, so Imma give my opinions publically. Also someone who is cheated in past, if it was exam, then gov bans such peeps fir re appearing in future, if it was in relationship, people will always see them as an unloyal person. It's not if you cheat again or not, the thing is if you've cheated in past,you will always be looked upon as a potential culprit in case things go wrong. It is 100% a possibility that you might have had nothing to do with it, but you can't erase your past.
Sure do.
And that's a pretty bold stand to take. Most of the Indians have cheated/showed blatant dishonesty sometime somewhere. India is one of the the most corrupt countries in the world. Good luck finding a non cheater for your 'good problemsetter'. I am sure even the other author has cheated somewhere sometime.
Nice to know about your Love for Graph Algorithms.
I've known Aryan since Grade 6, and his dedication to problem-setting has always stood out. I genuinely enjoyed the round — especially found Problem C quite interesting (though A felt a bit more time-consuming than a typical A ). The cheating issue is beyond the setters’ control, and it’s unfair to hold them accountable for that. I Appreciate the effort you put in, Aryan. Setting a full CF round and CodeNite is no small feat — you gave it your best, and that truly shows.
Keep going strong
I solved ABCD in this contest. The following comment is my opinion to problems in this round.
A: Maybe a little difficult. I think this problem has too more cases for d2A. However, just a little. It is a good problem if it is regarded as a hard d2A.
B: I have seen this trick so I think it is easy and suitable for d2B.
C: Excellent constructive algorithms! It is interesting and I love it.
D: This is more like a fill-in-the-blank problem in a mathematics competition rather than a programming problem. I'm not criticizing this question. What I mean is that it would be more appropriate for it to appear in the field of mathematics than in Codeforces. Of course, it's just my opinion.
E: I tried to derive the optimal case (essentially a proof of greedy), but it seems that I'm still one step away. I haven't seen the solution to the problem ( so when?), so I'm not quite able to comment on my feelings.
Anyway, I remain neutral on the problems of this round. The opinions on each problem are as above. No matter what our preferences for the competition are, we always hope that there will be more high-quality competitions. So thank you for your contribution.
Can we get some clarification on whether the round will remain rated or not?
will we get rating ffor this contest?
Don't be too sad— the problems still have a lot of good things. But there were really too many cheaters this time. I think this is about a person's personal character, not their country or race. We can't judge everyone based on a few people, because every country and people have their own greatness. If I get a chance, I hope to do your problems again next time. I also hope my rating will be higher then, so I can give a better review.
I think the problems were well-made, you guys did a great job. I also believe a part of the backlash came from the fact that many of the top pages of leaderboard were filled with Indian newbies. It is obvious that most of them were cheating. Some people seem to associate you with them, perhaps just because you're also Indian, which might have contributed to the hate.
Another factor was your co-author being caught cheating in two rounds. A clear statement or clarification from him could have made difference in how people gonna take that.
Yeah, we should participate for the sake of problem-solving, but rating also does matter to most of us, no matter what you say. All authors should do contest time cheat checking, like at least for the top pages as cry did.
Absolutely
I appreciate your response message being much kinder than the expert guy without as much explicit language. I feel like the way that he wrote made it seem very defensive as if he had something to hide.Furthermore while I do believe that innocent until proven guilty, I felt that Mindeveloped made some darn good points. My own take however do as you will
Dragmon would you mind explaining this?
I have, please see. Thank you for briging the comment to my attention.
Thanks for clearing it up.
Hi [aryansanghi], I noticed you’ve authored some really high-rated problems, That’s impressive! I’m curious — how did you develop the skills to create such tough problems? Would love to learn from your experience.
Hi, as you can see in the blog, key point to author high-rated problem is having a good grasp of even one topic that you can then create and solve high rated problems in. Also, I think it is different in solving problem under time constraint of 2-3 hr and creating problems where you give days or weeks of your time to solve the problem you've created, so even if they aren't a part of your high skill skillset, you can still solve them as you give enough time. Regarding problem ideas, you can refer to the blog where I've detailed how I came up with the ideas of problems and such.
I completely disagree with Um_nik's post if problem is solvable by AI within minute then this indirectly means your problem is not original. and non-original problems (problem from somewhere with different wording) considered bad and particularly bad if it appears in Div.1 or Div.2 beyond C or D.
There are many problems that are good, original, and fun to solve yet AI can still solve easily, personally disagree with this sentiment.
I'm really sorry for anything I said before.
I didn't realize that many other problem setters had also cheated in the past but have since been long forgiven.
I didn't really mind the problems either. It was just unfortunate that they were easily abusable by AI. But that's not the fault of the problem setters; it's the fault of the cheaters who use AI during the contest.
I think instead of writing
I don't care if people cheat or use AIyou should write justI don't care. That would be more accurate.And these
I don't cares are the main reason why you shouldn't be a problem-setter, not the rating itself. But to be honest, there is one more thing that is not fully on you. Every author loves their own problems. It's very typical, all of us are more biased towards types of problems they like. But, as I said a billion times, coordinators at Codeforces don't do a great job of problem selection and validation.And I want to explicitly respond to
I came to this platform to enjoy solving problems with a time constraint. This website is about not just math problems, but about programming problems. Which might involve math for sure, but when among the first four problems there is no programming required at all, I would say this has nothing to do with programming. So those who love programming significantly more than math do not enjoy rounds like this at all. Again, I think it should be moderated by coordinators, but never happens, unfortunately.That's because coordinators, unlike you, have taste.
Having cheaters as co-authors? I don't care!To be fair, it was once 2 years ago (in contrast to someone who is actively cheating), though I agree that this should have been addressed publicly before the contest.
Giving pure math problem for D so that everyone can just ask AI / friend / Google for a formula and you will never find out if they cheated? I don't care!As someone who isn't the best at mathforces, I agree with the first part (that pure math is annoying), but the second part falls under "I don't care if people cheat or use AI".
Problems from A to D do not require any coding, just math? I don't care!C was greedy, the math in B was minimal (with the main observation being that collisions don't matter), and same with A (though I do agree that it was a bit of an annoying problem)
The community disliked the round so much so that its post became one of the most disliked posts ever? I don't care!Considering both authors made a post about it, it's pretty clear that they do care about this.
None of the arguments you made prove that the authors "don't care". You can argue that some of the problems were sub-par, and that does seem to be the general sentiment, but I see that more as a flaw in the contest, not as a flaw in their character.
Hi Dragmon, I just came across this blog. I'm an Indian college student, and lover of Physics (especially Olympiad problems). I'm new to CodeForces.
I'm interested in the problems you set for brilliant.org, I didn't know the community did that! If you can, please share them, I'd love to take a look.
(I'm writing this as a comment because I could not message you, lol)