Inspirational! I can't wait to see where your resilience and consistency takes you next!!!

how about getting a GF?

so, you know how to get a generating function?

But in his case, it seems like creating an observer account to solve lower division problems is okay

implicitly yes, because you must use all the characters for each copy

omg kevlu8 round

what geometry knowledge is required?

not sure what u mean, there are lots of variety

TheScrasse is my favorite cherry.

max

results and upsolving will be released next week

I appreciate how dedicated he is to making these tutorials on top of being a Phd student in his own life. It must take up a lot of his free time. Moreover, I'm not sure if he's even getting any benefit from this (besides growing his youtube channel).

P.S. I do wish he officially participated in the contests he'd be reviewing though.

wah

your mother

Intellegent is just salty that he has lost 10 TLE duels in a row to reirugan.

As a non-tester, I will probably find the problems really cool! I encourage you to participate and have fun.

Actively solving problems on CodeForces

17 problems solved for all time

you missed low taper fade

skipped, should be punished shortly.

As an INOI testuwuer, glhf :D

yes

I third this. I used to be one of those authors that automatically disregard problems that GPT can solve. Of course, when only 4o was a thing, this was still realistic. Now, with o3-mini, I think it's a whole different situation.

I've probably rejected countless problems (either by myself, or by my fellow team of authors) just because GPT can solve it. I used to uphold this standard, but now I'm convinced it's not worth the time spent and the quality tradeoff. After all, lower division contestants are supposed to be educational, and I've always believed that. I don't think my fellow authors deserve to be frustrated when they propose a perfectly fine problem, but it can't be used just because of AI.

If you're an author, I know it can be quite frustrating when you spend so much time making a round, just to get hated because of complaints about AI. I'm completely in support of ignoring these complaints and make the round quality the best possible. I hope other coordinators and authors will agree with this.

<3

Participate for a fumo.

smashman

Leaving this comment so me in 50 years knows I was top 1 contrib during this time.

hi

I thought E was easier than D but chromate00 held me at gunpoint for the rating predictions.

USACO has much more participants.

You have to full-solve bronze and silver beforehand.

I didn't BOOK MY SEATS for taking classes for USACO (BRONZE,SILVER,GOLD,PLATINUM) our 2025 batch is open now, so I will fail.

lol this solution is $$$\mathcal{O}(n \cdot 2^m \cdot m)$$$ and it still (barely) passed.

Btw, your chinese contest might just have an unordered_map hash hack. It's probably not related to the judging server.

i have no idea how to get past 3%

perhaps

As a tester,

This only intersects with round 1 of advanced. Beginner round is much later after this contest ends.

... and bitmasks

soo who has a flows solution to 2009F - Запросы Светлячка?

Can you try different models besides linear regression? Would be interested in which model is the most accurate.

Remember me when you reach lgm :sob:

ok lil bro

how many alts do u got

are you the imposter?

As a testuwuer, I was surprised to find an ICPC contest with div.1 quality problems. Really high quality stuff!

American ICPC, step it up ;-;

nice edit lil bro

i sure do wonder why everyone is picking rock

bruh

not you again

cope and seethe

play hypixel

ugh i failed usaco dirt. the first problem, "even cows", was about finding if $$$N$$$ was even and the limits were $$$1 \leq N \leq 10^9$$$ and it was so hard. my first idea was to store an array of all even numbers in the world but it said memory limit exceeded for some unknown reason. then i realized exactly what the problem was. i was using to much memory!! so i reduced the array size to $$$10^5$$$, and it passed the sample test case :). but for the rest of the test cases, it had an exclamation mark for some reason. i must have not added enough for loops and recursion. so then i added infinite recursion and resubmitted, and was sooo disappointed when it time limit exceeded on the first test case :(. but at least the code ran, unlike when it was just a big fat exclamation mark, so it was an improvement i guess. but right as i was about to submit again, the 4 hr time limit passed, and i ran out of time. does anyone know how to solve this problem??

You can message me to be considered, but I usually only accept / reply to people that seem active, promising, and honest on a variety of factors.

guys i think we got a bit carried away while writing this contest

As a tester, hackerrank is kinda mid

same, my internet provider blocked the website as phishing website

You can use this amazing command-line tool by cryan.

masterpiece

Thanks, good to know. We would never have found it on our own D:

I can't access the link. Can you send an image?

Yeah, we're well aware that most of the correct submissions from rated participants are probably from GPT o1. This does explain how G got a lot more solves than F. Our testing proved G to be a lot more difficult than F, especially when the technique in G shouldn't be "standard" for div.3 participants.

I can't really do anything about it though. I did ask admins to liveban some cheaters, but there is way too much to manually look through. It's up to the CF plag system.

My bad, I wrote the note and I'm not really familiar with python. You'd have to use a frequency array of length $$$n$$$.

Obviously, you can binary search for the first $$$i$$$ such that $$$f(1, i) \gt 0$$$, but it's still $$$O(n)$$$ queries in the worst case.

I'm sorry that Codeforces Round 979 (Div. 2) gave you so much agony, but keep at it!!

I FSTed both A and E1, and my B code RTEd on main tests. gg

top1codeforces is clearly using GPT. Why is he still getting first solves credits on A?

no

what's the point of having gpt implement something for you when you can't use it to implement something in contest :icant:

Let's just pretend that Natlan didn't come out two months ago.

no, proofs are too hard, i think he’s better off improving at guessing

Also, opening a submission link on a browser where you’re not logged in (e.g. discord on mobile) is completely not allowed. It’s really inconvenient D:

cat

aventurine reference?!?

I did $$$O(RC \log^2 RC)$$$ using 2d BIT and binary search.

Partition into multisets {0, 1} and {0, 1}, score is 2 + 2 = 4.

shameless plug

unusual starting time so people have time to eat lunch before meta hacker cup

It's not possible, unless you have access to the problem's polygon form and you can edit the scoring from there.

wavelets got a 1600

I thought everyone learns to put cin.tie(0) -> sync_with_stdio(0) in their code the first day they start cp.

Guess not :/

Thank you for sharing polygon materials! For future setters, can we normalize this?

I don’t mind standard problems. In fact. the last div.3 I set were all textbook problems (according to Um_nik). However, where I draw the line is when the problem feels like knowledge-checks. H is a good example of it. I don’t think the idea is difficult whatsoever, but knowing xor-hashing is the only bottleneck that makes it H. This is the main reason why I was against the problem.

The same can be said about the dijkstra problem. The same problem (with almost no modification) can be found in the first few listed problems in the shortest paths section of usaco.guide. The hardest part of the problem is implementing dijkstra (and knowing that it exists).

Hi RobinFromTheHood, idk if you know me, but I’m one of the testers for the original set (when frugal was still in the mashup). I tested through Vlad directly.

I recall providing some harsh criticisms regarding the set. In my feedback (which I DMed directly to Vlad), my feedback included the following:

• Remove frugal version. I didn’t see an easier way to approach this problem than the nonfrugal version.
• That dijkstra and tree dp were way too standard even for Div.3. I didn’t explicitly say that they should be replaced, but I hoped you guys would see the issue.
• Regarding the last xor hashing problem, I received a message from Vlad to test it before it was even prepared. In my feedback, I remember saying that it’s not a good problem and it should be replaced. I even provided an atCoder problem which is really similar to it. To my surprise, the exact same version ended up on the actual contest. It’s almost like my feedback didn’t matter at all. At the end, I also proposed a modification of the problem on a tree. I received no response from Vlad after that.

Since we were not communicating directly, I’m interested in how my feedback reached you. Were you able to receive all of my feedback, including criticism, directly from Vlad? I’m especially curious about the third situation which I listed. Did you guys receive my modification and think it was not necessary, or didn’t receive it at all?

SecondThread

No, but I'm not sure when it doesnt crash. When I read the error logs it's like title too long (lol??) or some weird error with statement

This feature has never even worked for me. It always loads some weird error but preview in HTML works fine.

If you can’t understand G2 or G3, you can pretend those problems never existed, and move on. Alternatively, you can note their existence down and return later. It’s up to you.

A-G1 already makes a complete div 4.

"Also, I said I expect the target participants to be able to SOLVE or UPSOLVE the problems."

Ok? isn't that the purpose of G2 and G3? For div.4 participants to upsolve? Of course, you don't have to upsolve it right now. You can always come back when you're better. The problem doesn't disappear.

I don't see any harm in contests with a harder final problem than its intended division. It gives strong participants of the intended division something to think about for the rest of the contest (instead of doing nothing), and makes more of a competition for higher rated participants.

ahh oops, thanks for letting me know

antihack is crazy

yunli will come home

Auto comment: topic has been updated by cry (previous revision, new revision, compare).

...but the whole point of div 3 and 4 contests is so lower rated participants get higher delta because they aren't competing with stronger participants?

bros a self-proclaimed tester

are there original problems lol

wait im not the only out of competition participant to mald and bald on B!?!?

fire logo tbh

Thank you for writing quality stuff recently! I hope you get into top contributors soon :D

The standards for div1+2 rounds and div2 rounds are different. Not all hard problems in div.2 can go in div. 1. Problems that are deemed standard for div.1 participants may be educational in a div.2 round.

As a tester, I agree