#ai-village-capture-the-flag-defcon31

1 messages Β· Page 7 of 1

minor falcon
#

can we get a flag if we bet correctly the 24 flags of the top 5 ?

ember relic
#

real nice

#

been working on granny for two days trying to match the models

#

and now i notice im not using pretrained weights

ornate marsh
#

did last year's participants also have a discord channel to commiserate with one another?

ember relic
#

yes

sturdy gorge
#

Even loading the weights i cant match the models 😦
Tried more than 30 at this points

#

seems like bruteforcing the api is gona be the way

#

really eager to see the solutions this year, to have a chance to actually try for a medal next year

ember relic
#

tbh i cant even match the pytorch pretrained ones with the keras one πŸ˜‚

random minnow
#

"i cant even match the pytorch pretrained ones with the keras one" of course you can't. they are not the same one

#

in the first place, waht makes you think they are the same? second you can easily check if they are the same or not. lastly, google can esaily help

ember relic
#

i dont know, i guess im not touching grass enough these days

sturdy gorge
#

tried keras, torch and kaggle models, no match πŸ˜”

brave briar
#

has someone reached a perfect match ?

#

not sure it is possible, they also could do an extra train batch

sturdy gorge
#

looking throught the chat several people have claimed a match

#

in my case, not even remotly close tho haha

ember relic
#

it shouldnt be a match right out of the box

#

but what worries me is do i have to try out all of the models to find out which one is on the api?

glass bay
ember relic
#

i know what model to use but not what weights

#

given that there are multiple pretrained ones

glass bay
#

assumed it to be the default and that it is the only one

ember relic
#

so far i've found 3 official versions

#

since keras and pytorch dont match

brave briar
#

keras and pytorch don't match after preprocessing ?

ember relic
#

i didnt check that

#

i checked only the weights

#

is the preprocessing different between the two?

brave briar
#

yep

ember relic
#

damn

#

didnt know that

#

thanks

hard perch
#

Is pickle an LLM? That seems the only plausible thing considering no actual code executions. Surely the most frustrating problem

past brook
ember relic
#

yea but thats not my concern

sturdy gorge
#

cant understand what im suposed to do on pixelated

#

OCR is working fine on whatever text i give it, just no clue what to give it

light mulch
# ember relic i didnt check that

There was some interesting discussion already about preprocessing. I found the comments by hengck23 very interesting. But I have to say I also have not found the model yet. harold

ember relic
#

whats bothering me is i dont know if every pretrained model (keras, pytorch imagenet v1, v2) will work if i find the right preprocessing

topaz ember
glass bay
#

yea that's getting hinty

ember relic
topaz ember
random minnow
#

"The host said somewhere in this chat, that the difference between the mobilenet implementations is not expected to contribute much."
i don't recall reading that. which exact message by the host are you refering to?

#

image adversial attack are divided into black box and white box. you should google about them

random minnow
#

be careful. i would only say, there are methods that require for you to know the model (+ other processing) completely and methods that doesn't (complete no knowlege). there are in-between methods as well (i.e. partial knowledge). in short, google more to find out what methods you want to use and the necessary conditions to make them work

topaz ember
#

yeah, everyone is free to make their own interpretations of what host says

#

more knowledge is better of course

mild shale
light mulch
#

Finally got Granny1 πŸ₯³

#

Congrats for place 1, Yuri πŸ˜‰

queen garden
#

Insane ! Good luck on the last 2 πŸ™‚

craggy beacon
final path
#

Wow, that's very impressive! @craggy beacon

pallid trench
#

any hint on granny 1 pls.

light mulch
#

read this discord channel πŸ˜‰

modern anchor
#

Which ones left for you @craggy beacon

sturdy gorge
#

Any resources for pixelated? Really cant understand where to go, i can get the OCR do detect any text, just no idea what to give it

final path
minor falcon
#

I officially give up, silver is not that bad πŸ₯²

#

Good luck for the tryharders of the last week!

gaunt anchor
#

I am happy with silver also, a gold will be better (I need a solo gold) ... it seems so close to achive yet so far ! I hope I have it to go through this another week !

outer sundial
lyric token
#

GMaster

gaunt anchor
#

for grand master ... but its a gold in the end

#

And I am so close yet so far :/

minor falcon
outer sundial
#

After reaching 22 points.

fallow valve
#

Congrats @craggy beacon
Hard to keep up withπŸ˜‚

sturdy gorge
#

desperatly trying to get granny 1 and 2 for a measly bronze
alas, no luck

brave briar
#

granny 1 is so slow ... It s such a pain !

rocky jacinth
light mulch
#

Only one more for me then. Might be doable πŸ™‚ Semantle 2 at 0.97 and some ideas for pickle and pixelated.

past brook
#

Im 20 but im trying for silver

#

I got all the annoying ones left. Granny 1/2 are the most likely ones for me

#

Maybe inversion

#

Im not sure CIFAR is even worth trying from what ive seen in the chat

sturdy gorge
cloud prawn
#

I've got to ask because it's been killing me. @gusty warren When you made this comment was around the time when you solved passphrase. I now and realizing you might have been referring to something different all together. Am I reading too much into it?

thorny ivy
cloud prawn
ember relic
thorny ivy
#

Can you imagine if there are people with 25 or even someone with 26 waiting for the last day, last minutes to submit? Lol

final path
light mulch
minor falcon
#

not valid now that we are close to the end. Someone that would have had 25 flags keeping it secret would just have lost the first place

light mulch
#

I meant of course submitting but not telling in this chat what he solved...

ember relic
#

what is happening with the granny endpoint?

#

its never been this slow

minor falcon
#

ah sorry, yes keeping the information secret is indeed a good strategy at that point πŸ‘

light mulch
#

I mean he posted it for like 5 seconds but then wisely deleted the message πŸ˜„

abstract rose
minor falcon
#

all those people (edit: made a mistake and included some people from the club 23, mb)

brave briar
#

hey @olive ledge or @limber flower, may you have a look at Granny 1 ? It seems it is down 80-90 % of the time. Thx

ember relic
#

not sure what changed

sand veldt
light mulch
#

now i feel the OCR pain of the previous posters 😀 πŸ˜„

final path
minor falcon
#

next 24 should indeed secure a prize, just because its unlikely that another 23 will get 2 flags in such a small amount of time I guess

rocky jacinth
#

I am surprised! Seems like WTF5 got fed up with our conversations and spat out the flag.

minor falcon
#

i am not sure, there is probably exceptions, but from what we can interpolate from the people and discussions here, appart from a few outliers that solved inversion or passphrase, most of the people with 22 flags are stuck on passphrase/inversion/hush/cifar/grany3.

#
  • it was more a joke than something else πŸ˜…
grave frigate
#

πŸ˜‚πŸ˜‚

grave frigate
gusty warren
severe pasture
#

new hint just dropped pog

grave frigate
gusty warren
#

TBH though, I may or maynot be misleading you all unintentionally. My solution is quite stragith forward and I believe many people should have tried the same thing. I am very surprised that many people still can't solve passphrase, I am starting to suspect that I did something different that I am not even aware of...

grave frigate
minor falcon
#

You might also have got all stars aligned at the right moment :p

cloud prawn
#

Thanks. That at least confirms some of the things I've been trying. But passphrase has been really tough for me and I've gone down many dead ends. At least that confirms my suspicions about your previous comment.

gusty warren
minor falcon
#

I wish i got it on inversion ahaha

gaunt anchor
#

imagine a lottery ticket in granny3 ....

minor falcon
#

i think for passcode a lot have been said also in the past that could mislead many people

#

granny3 i have a last idea that i could have tryied, but it implied mounting a cluster of server to query more efficiently :p

craggy beacon
minor falcon
#

i think the luck is the thin layer that allows you to not wrongly deny a direction because of a dumb error/missing info on some kind of postprocessing

#

pickle is a very good example of it: you miss the sweat spot, you can end up into a hell of complexity

violet trellis
thorn quarry
#

Does the static flag in inversion mean anything or is it just a random letters?

sand veldt
minor falcon
#

what i need is not cpu/gpu time, but different ips :p

sand veldt
#

ddos could cause ban LOL

minor falcon
#

when you run multiple kernels on different tasks (or on the same task with different seeds/approachs), you do exactly the same

sand veldt
#

well, it's a matter of scale

random minnow
#

"what i need is not cpu/gpu time, but different ips :p "
you need to reduce your search space by working on the most probable pixels

minor falcon
#

yep I know, but I would have tryied another way that I think could be promising giving my current view of the problem but that I will not implement in any case because I am too lazy :p

sand veldt
#

hope granny3 is not the same as cifar, which can only solve by api call

minor falcon
#

? cifar is the exact problem that cannot be brutforce

sand veldt
#

yeah, but everytime need to call api to check

#

can anyone optimize cifar locally?

hard perch
random minnow
#

there is a new kaggle competition

#

so maybe there will new attack for next CTF: fool LLM text detector

thorny ivy
hexed plover
#

has there been any clarification on passphrase? I managed to target specific scores, but I didn't get any flag.

ember relic
#

there have been some hints in this chat

#

i havent even tried to solve it so i have no idea if theyre useful

#

take a look here

hexed plover
#

Thanks!, i'll take a look

viral bough
#

No idea what to do

viral bough
#

I got IP1 πŸ™‚

#

and IP2, yay!

#

just messaging here gets me closer πŸ˜„

sand solstice
#

IP is baffling at first, but you learn a bit and both fall pretty quickly together

acoustic temple
minor falcon
#

looks like the 5th spot for money has been taken. Good luck to the people targeting the 25 flags now!

gaunt anchor
#

there is still a week ... you never know ... for CIFAR and Passphrase its once you see it you get it .. so the flags for both are floating in the air ...

gaunt anchor
#

congrats @severe pasture

#

no I didn't I tried in two diffrent domains (or directions) ... nthn so far

#

everytime I think I have something that fit ... its try again ... CIFAR could have logically different answers ... but there is only one answer will get the flag :/

random minnow
#

"looks like the 5th spot for money has been taken. Good luck to the people targeting the 25 flags now!"
the doctor cannot save all patients.
we cannot solve all problems.
that 's the reality of life...

we can only try our best and there is always the next time

gaunt anchor
#

there is still a week , then break ( I hope) till the santa challange next month !

#

however the llm detecting ai generated text seems intersting ... no break πŸ˜„

past brook
#

is there any forum or website where you can find AI security competitions?

#

I really enjoyed this one, but I stumbled upon it on accident πŸ˜†

light mulch
final path
unborn sail
sand veldt
#

cifar is actually fun, guessing other's mind. like guessing other's ins password

olive ledge
abstract rose
minor falcon
#

We'll all know in a week

light mulch
modern anchor
#

Semantle2 done finally. Was stuck on 0.98 for a long time

ember relic
#

finally matched the granny 1 model

#

god am i stupid

gaunt anchor
#

you are not ... you just didn't see it

#

A week later when we see how CIFAR is solved ... we need to say "oh ... we didn't see that" ... its better than god I am stupid for wasting daaays

sturdy gorge
#

aheems

#

still have not been able to match it

ember relic
#

im also somewhat weirded out by this model matching

#

because i would have sworn that i already tried what turned out to work

#

i guess revisiting old ideas when youre well rested makes sense

naive umbra
#

I was able to match the model yesterday. Hopefully I can finish it tomorrow.

sturdy gorge
#

starting to think the problem is not the weights but the preprocessing zzz

past brook
#

im gonna try model matching aswell, cant get black box to work though there is probably some way around it

past brook
hard perch
ember relic
#

im not sure if im allowed to say hahaha

hollow coyote
#

MNIST: 'Try again!' 😑

#

its so hard to shake off the idea I attempted because I feel like its the only thing that makes sense. Got myself stuck in some local mind optima

limber flower
ember relic
#

FINALLY

#

CLUB 23

#

LETS GO BOISSSSS

thorn quarry
#

CIFAR 10 or CIFAR 100?

final path
#

who knows πŸ€·β€β™‚οΈ

random minnow
#

"CIFAR 10 or CIFAR 100?" the answer is the 7 th part of the inversion static flag solution and 8th part of the hush static flag solution

amber sage
#

sloth

light mulch
gaunt anchor
#

I happy with sliver ! ... I ran out of ideas for the last four ... and I am exhausted :/

orchid sundial
light mulch
#

Finally i got inversion! πŸ₯³

#

was really fun

lyric token
abstract rose
light mulch
ember relic
#

i see youve solved it as well, nice

light mulch
#

What do you think is a good candidate to work on for 24? I havent yet tried CIFAR at all but what I read, it seems impossible.

final path
#

passphrase probably. there are some recent hints in a chat, may help

ember relic
#

i had way too many logical ideas for cifar but none seem to work

light mulch
#

Is there some evidence for solved Granny3, Cifar or hush in this chat?

ember relic
#

its more of you either see it or dont

ember relic
light mulch
#

yesterday, or so, some 24 on the leaderboard was renamed granny3. But probably trolling or? πŸ˜„

ember relic
#

waaait

#

weve got another 24 on the leaderboard

light mulch
#

yes πŸ˜„

#

but there is still room for 24 gold πŸ˜‰

ember relic
#

thats what im hoping for

severe pasture
mild shale
#

Hi

#

everybody

#

its good to rejoin the party

cloud prawn
orchid sundial
cloud prawn
#

Nice work!

naive umbra
#

0.9999921321868896, 'Granny Smith' please tell me I'm close

past brook
pearl pawn
#

I cannot solve MNIST and CIFAR. I can't believe I can't even count.

naive umbra
#

Got 1.0 prediction for granny smith but no flag...

#

and I got the flag

#

finally

#

And granny2 is also solved. Finally I can enjoy my weekend haha

gaunt anchor
#

I miss old days .... what old days ? ... when flags used to show up 😦

final path
#

17 days ago, as for me harold

ember relic
minor falcon
#

i was checking my old notebook, my brain is fucked up, I get a dopanin shot when I see a output starting with {"flag":"gAAA.."}

#

are you guys going to other comps after this one ?

acoustic temple
mild shale
sturdy gorge
ember relic
unborn comet
#

Was anyone else getting sporadic 504 timeouts for WTF? WTF3 in particular?

random minnow
#

this is my sole opinion and maybe most would not agree. while the CTF tasks could be made more difficult, i prefer team work and also discussion. we all can go further+faster in a team and discover more security bugs . it seems that everyone is stuck.

craggy beacon
#

but we have learned more that way because you can't delegate some stuff to others

nimble matrix
#

how can I convert the image or tensor to base64 string for granny?

nimble matrix
#

@ember relic Thank You!!

sand veldt
#

{"message": "An unexpected error occurred in scoring script. Check the logs for more info."}, I wish I can check logsharold

thorn quarry
#

I'm trying to understand the connection between sentiment scores and the passphrase(password) concept but no luck so far

olive ledge
olive ledge
#

Didn’t see anything particularly bad, service is still up. Check your inputs πŸ™‚

nova pelican
#

For those with 20+ {"flag":"gAAAAABl==GG-YourFire!"}

lost relic
#

I went to the shower and I have an idea for passphrase, no joking πŸ™‚

fervent obsidian
#

Is this relevant to something you see in a shower environment? πŸ™‚

abstract rose
#

Inversion endpoint looks bad: "{'message': 'The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.'}

waxen lynx
#

Inversion is unstable since friday to me.

abstract rose
spice hawk
acoustic temple
ember relic
#

i had an idea for passphrase but its really hard to find a sentence with the same sentiment while keeping in mind the constraint

nova pelican
#

Where's Grok @ with all its math might.

timber lake
#

Why does hush sometimes returns 2 elements whereas sometimes 12?

hollow coyote
amber sage
#

still, I cannot count.

tribal plank
#

I wonder how simple it is.

pearl pawn
queen garden
amber sapphire
#

Finally got Mnist...can't believe was so simple 🀯

mild shale
timber lake
#

oh sweet semantle! finally

naive umbra
#

i'm at 0.97 for semantle2. I'm hoping I can get it before the comp ends.

eternal island
#

@naive umbra check this out

naive umbra
#

Got the flag for semantle2!

acoustic temple
#

so close and still so far in Inversion, I can't understand what is it, and also can't give up at the same time, strange feeling πŸ˜…

hollow coyote
rocky jacinth
queen garden
random minnow
#

just a quick question. i am also taking part in other kaggle competitions. but i find that their discord channels are pretty quiet.
am i right to say that that is the norm? and this discord channel is unusually active?

devout jasper
#

because we suffer together

#

πŸ˜„

olive ledge
#

IT WAS THE FRIENDS WE MADE ALONG THE WAY

minor falcon
lost relic
nimble matrix
#

Why is my shower not giving me any flag

thorny ivy
#

just kidding... or maybe not πŸ˜†

valid cobalt
random minnow
#

"Why is my shower not giving me any flag"
you took the shower at the wrong date

#

time, place, person

minor falcon
#

maybe we can build a model to find optimal shower parameters to get ideas

amber sapphire
#

Keep oscillating between "not so dangerous" and "to much dangerous"...

lost relic
light mulch
#

Wow! Congrats to 2nd place!

tribal plank
#

ε•ŠοΌŸ

fallow valve
#

I think two with 25 is enough thoughπŸ˜… πŸ‘€

tribal plank
#

I think 26 is possible.

#

Apple should be the most difficult one.

craggy beacon
#

because only one unsolved is granny3 pbly

tribal plank
#

Because of the existence of easter egg.

past brook
ember relic
#

it doesnt work most of the time

cloud prawn
ember relic
#

whats up with this message

#

i got the same

glass bay
#

imo the challenge creation process for granny3 was as follows:
||get an image, do stuff to lower accuracy, add this god forsaken pixel, lower accuracy more, remove the pixel||
and if that happens, ||imo, there is a solution, but it is somewhat degenerate in the sense that it would be out of place in the reference to the original, or after-second-lowering image structure, thus leaving it be in a part of search space that has no locality or continuity or something||
'cause otherwise the answer would've been found already by many people since the papers' algorithm ||uses locality and continuity that are coming from image structure, that may have been destroyed during the perturbations and|| may be useless here, requiring new approaches

final path
minor falcon
glass bay
minor falcon
#

why not ? The pixel attack can be performed from the whole pipeline of the model, including the preprocessing

#
  • it has been said by the organiser that a match of the model is not necessarly needed (they didnt use it to test the challenge)
glass bay
#

then it would have worked for someone already

#

certainly the wolf image has been artificially altered for that exact model to give 28% acc because of the way that it is recognized by other models with 80+%

tribal plank
#

but I still do not know why this processing prevent differential evolution

minor falcon
#

For me its just that there is a lot of local minima, hence its very difficult to optimize.

past brook
#

meanwhile im not even able to solve granny1 with black box attacks πŸ˜‚

minor falcon
#

I think I know why DE is not working well and I have a suggestion of algo to try for granny3 that I am now lazy to implement and that would take a lot of time and calls, i'll be happy to discute it once the competition ends

queen garden
#

I have some thoughts on g3 as well but will wait for competition end to share, as should everybody πŸ™‚

glass bay
#

there are alternate algos indeed yet imo DE should work best

#

that's the thing i was trying to say above - that global minima may be in an unfindably small region and have nothing promising say 1px to the side and 10 units of brightness lower/higher and no way DE or something like that will find that

pearl pawn
#

Finally I got MNIST flag. I didn't expect this answer.

sturdy gorge
severe pasture
#

another 25 πŸ‘€

gaunt anchor
#

its going to be a looooong three days ! πŸ˜„

light mulch
cobalt flint
#

pickle is just so soul crushing... doesn't make sense to go for any other challenges since pickle is apparently so much easier than them, and here i sit having sunk 20+ hours into trying to be dangerous by sending pickles to a server where no code is executed... and no feedback is given. at least for mnist we know we have to count something. really not a fan of challenges where the goal is not given. rant over

jagged sluice
#

vinegar + salt + sugar + cucumber + spices

#

stick in jar

olive ledge
#

Simply having a pickle file that executes code ISN’T necessarily dangerous.

#

For the same reason you can’t email PE files into anyone with even a smidge of security.

tribal plank
#

I solved pickle really by thinking how waving keyboards works...

#

And I still do not know why my solution is accepted.

cobalt flint
olive ledge
jagged sluice
#

attach to pole and hang outside

#

boom

#

flag

gaunt hollow
#

btw I have no idea on MNIST at all lol

soft lintel
random minnow
#

my idea is to let chatgpt to read all messages in this discord discussion and kaggle forum

#

then we can make a kaggle CFT tutorial app

random minnow
#

the important think is the "chain of thought" : how the winner thinks of this so0lution? and this is waht most people want to learn

#

we can ask winner to contribute their chain of though, solution code experiment results, what worka nd dose not ...etc . all content contributer for the app get gat paid roalty.

ultimately we want an app to give hint (not full solution) to guide us step by step

jagged sluice
#

so just gpt4

#

but finetuned?

olive ledge
#

Prompt-Engineered. .

jagged sluice
#

time to make AIV-7b a thing

fervent obsidian
#

all those pickles will be lost in time

#

just like my tears in the shower

tribal plank
#

The legendary shower should be encapsulated into an AI model

minor falcon
#

talking about gpt4, there has been an update, the current knowledge cutoff is now april 2023

acoustic temple
#

is passphrase really about to having a sentence which return 0.3333 for each sentiment?
I can't understand what question asked for

#

"a difference sentence where everything is equivalent"
is it sentiment score? sentence meaning? none or both ⁉️

random minnow
#

why his shower works

olive ledge
past brook
light mulch
#

Just got an idea for passphrase, but did not shower. So it's probably wrong. πŸ˜„

queen garden
#

Omg, finally got passphrase. No shower needed

ember relic
#

congrats

light mulch
#

Congrats! What changed since this "ancient" message? The recent hints? πŸ˜„

queen garden
ember relic
#

idk how you can start solving anything before you solved cifar

#

im too pissed off to start anything else hahaha

light mulch
#

Cifar is too frustrating. Twice a day I think I finally have the solution. Code it out for 10 minutes and only get "Try again!"...

gaunt anchor
#

I have a final idea to passphrase I going to try ...

acoustic temple
ember relic
#

i havent solved passphrase but just a heads up

#

in case you didnt see there was a different prompt for passphrase in the first few versions of the notebook

#

myb it helps

light mulch
#

are you refering to the ctf-starter-aiv notebook?

hexed plover
#

i appreciate a lot the effort on developing this ctf. But I really don't see any value on these obscure tasks. It is better to have a hard objective than an easy objective that with unclear instructions, it is just frustrating 😦

acoustic temple
light mulch
#

I also cannot see any difference. Version comparison on kaggle does also not show differences in passphrase

ember relic
#

i have some kind of different prompt in mine locally

light mulch
#

can you share the prompt?

buoyant frigate
#

They republished the notebook after some day, probably something wrong got there in the public notebook.
@olive ledge , is that correct?

acoustic temple
#

Maybe the challenging part is understanding it. Whenever we understand what is the objective, doing part should not be that difficult

waxen lynx
buoyant frigate
#

yeah, there was a change in variable name, but the prompt itself seems to be the same. Am I correct?

acoustic temple
gaunt anchor
#

Ok my latest and maybe final passphrase test is running ... let see the flag pleeeeaaaseee

olive ledge
#

Don’t over think it.

gaunt anchor
olive ledge
#

Hallucination?

ornate marsh
olive ledge
gaunt anchor
ornate marsh
#

I only want one more flag, so i'm jumping erratically between all my unsolved problems and failing spectacularly at each πŸ˜„

past brook
#

I just want granny, ive put it way too many hours not to get it

#

even if it wont change my ranking

ember relic
timber lake
#

And then there is β€˜I haven’t failed 1000 times. The light bulb was an invention with 1000 steps.’

gaunt anchor
#

No flag so far … but I think I am on the right track .. but I need to speed things up , I am so sleepy so I will update the strategy on the morning ( unless a flag shows) hhh

#

This flag means gold so it,s better show it self 😎😎

past brook
#

hmmm.. people are still getting flags at the 20-threshold

#

i might fall out of bronze lmao

olive ledge
#

I won't believe it until I see it!

ornate marsh
#

I'm trying to claw my way into bronze πŸ˜„

final path
#

Stumbled upon this today: http://www.paulgraham.com/top.html πŸ™‚
"I realized recently that what one thinks about in the shower in the morning is more important than I'd thought."

past brook
#

ahhh thats why I havnt gotten any flags

#

I havnt showered in 30 days

viral bough
#

I was able to load skops for cluster 1, but is there any tips on predicting with the data? It's probably simple but I'm not too familiar with skops

#

I did some basic searches on Google

#

(nevermind looks like it has the usual predict_proba and I just need to figure out how input it right)

gaunt anchor
#

My new idea failed me with passphrase ... 😦 .... there is one last edit I want to try ... then I am out !

lost relic
#

Hi @olive ledge , Is granny3 fine? I see many 504 status code on my last optimizations

tribal plank
#

I should have quit 12 days ago.

limber flower
gaunt hollow
#

Weird...Why the default code in pixelated challenge cannot run in my terminal, resulting in TypeError: Object of type bytes is not JSON serializable?

#

Sorry if that's a silly question. Does anyone have some clues?

lost relic
tribal plank
#

append a .decode()

gaunt hollow
#
import requests
import base64

with open('/kaggle/input/ai-village-capture-the-flag-defcon31/pixelated/reference.png', 'rb') as f:
    input_data = f.read()
    input_data = base64.b64encode(input_data).decode()

def query(input_data):
    response = requests.post('http://pixelated.advml.com/score', data={'data': input_data})
    return response.json()

query(input_data)

will return {'message': 'Invalid input.'}

tribal plank
#

data= => json=

gaunt hollow
#

Oh I see. Now Kaggle works normally

#

somehow the terminal run normally but outputs nothing = =

#

Thanks, that may be a bug in my computer

tribal plank
#

query(input_data) => print(query(input_data))

gaunt hollow
#

oh what why

#

it works

#

thanks many much!

#

why?????

granite goblet
#

Is Hush slow for you guys too?? Its taking like a minute now to query

faint mulch
#

I have the same question. Mine is also far from 1

ornate marsh
amber sapphire
#

Just solved pickle...so much time wasted...😩

#

Now hurry up for 20

neat light
#

omg I still don't get what I need to do for Pixelated...

amber sapphire
naive umbra
modern anchor
#

I need pixelated too. Wasted too much time on that

#

I know what text to enter but the OCR is so bad.

faint mulch
#

Any tips for Cluster 1? It's 3AM where im at, and im losing my mind trying to figure this out. 🀣

neat light
modern anchor
modern anchor
amber sapphire
neat light
#

haven't found anything useful here or anywhere else about pixelated...

sullen forge
#

use one same prompt solved all wtf1-6 except 4, looks like it can't break the securityπŸ™ƒ

acoustic temple
#

is chatgpt down ? I can't open my chat histories

fallow valve
faint mulch
faint mulch
# timber lake Read the hint again

the one that says "find which subpopulation is being systematically misclassified" or the one that sounds like a harry potter line 🀣

timber lake
#

@olive ledge semantle2 giving β€˜failed to process input’

#

ChatGPT is also down

amber sapphire
thorny ivy
acoustic temple
faint mulch
gaunt anchor
#

@olive ledge @limber flower The hush server is almost dead ... can you please check πŸ™‚

timber lake
faint mulch
#

Finally solved Cluster 1, after getting some coffee and burrito

olive ledge
jagged sluice
#

moo have you recieved the full printed mnist cifar dataset yet?

#

i spent extra to fedex πŸ‘Ί

gaunt anchor
#

Yesterday I said my last try ... then I am out ... can anyone just tell my head that I am out ...

gaunt anchor
#

I am not :/ ... I am just staring at passphrase .. hush ... and cifar ... granny3 no need to stare

#

granny3 is crop2 (from last year) to me

runic trellis
#

Are there any Cifar hints?

icy notch
#

try not to go crazy while solving it

lost relic
#

Are kaggle notebooks working for you? The code does not load for me 😦

surreal lantern
#

same here... I've been unable to load any notebooks for the past 2 hs 😞

acoustic temple
#

today nothing is working, kaggle, chatgpt, ....

lost relic
#

Excelent timing harold

past brook
#

Appearently it may take a couple of days before its fixed πŸ™ƒ

icy notch
light mulch
#

or gained admin privileges to gpt by uploading a image.

tribal plank
#

mischievous

devout jasper
#

wish I had one more week πŸ˜„

glass bay
#

I figured out an approach to hush...
But it requires 660 petabytes of ram...

neat light
#

getting a new output from hush....anyone got this one at some point?

glass bay
#

i had it sometimes, it was gone after matching properties of a file to send with the test noise

#

@olive ledge could you post the granny3 pixel right after the end of comp and leave the servers running for 3 days after? i reeeeeeally want to check if that needle in a haystack is findable and how close i got after wasting 2 weeks on it

neat light
#

anyone ever got this out of pixelated?

neat light
minor falcon
#

i am redoing granny 2 for a notebook I intend to put public.
So nice to see the probability going above 0.00069

#

I missed that feeling

jagged sluice
#

when people still haven't realized that you can pixel shift the entire timber wolf with a checkerboard pattern and it will pass granny3 inspection

minor falcon
#

actually it would be really funny if granny 3 was actually not an advertial challenge, but a challenge to bypass the 1 pixel checker

jagged sluice
#

the thing is that it is

#

and there is a way

#

i'll reveal in 2 days B)

minor falcon
#

25 hrs to be accurate

devout jasper
#

I hate inversion so much

#

and I'll be so pissed, I already know it

minor falcon
#

i feel your pain, inversion marked my game over here 😒

devout jasper
#

two positions shy from silver medal

ember relic
#

Im so glad this will be over soon

past brook
#

I really want to see a black box solution for granny1

#

ive implemented 3 different papers without results

#

i must be doing something wrong

icy notch
#

my black box solution for granny 1 work

past brook
#

There is one black box approach I havnt tried that I believe might work but I cba

minor falcon
#

if you want to experiment still, go on granny2 which is much less used, so has better latency

past brook
#

I got my mind on CIFAR instead, though its even less likely ill solve that one

tribal plank
#

I also checked if there is a color that granny may misrecognize because in the easter egg picture she wears a pair of sunglasses🀣

olive ledge
obtuse frost
#

I am getting a 0.94 score from Semantle 2. How (ir)realistic is it to complete the challenge by the competition deadline?

obtuse frost
jagged sluice
#

21 hours is plenty

#

time to shitpost for 21 hours straight 😈

ember relic
ember relic
#

Theres some sleep to catch up on

tribal plank
faint mulch
#

any hints for Cluster 3? Few more hours left until competition is done, but my brain is fried because of Cluster 1 πŸ˜…

tribal plank
#

Right way. So it is a big hint.

past brook
#

i would suggest deleting your message

amber sapphire
#

Sorry, I thought there were problems, I'm not so expert with CTF

gaunt hollow
#

I know this but no idea what to do next (to prevent accidentally giving hints I could not list my attempts though 😦 )

faint mulch
#

anyone? cluster 3? 🀣

tribal plank
#

Hint: play more with Cluster 2.

#

🀣

faint mulch
#

maybe need more coffee

glass bay
#

for cluster3: try multiple times. There may be some reordering happening from time to time when you try to order the clusters

#

it took like 20-30 retries to get it

outer sundial
tribal plank
#

This is why i am serious in suggesting to play more with Cluster 2

minor falcon
#

I agree with Roy, with the right pre-processing, Cluster 3 become straight forward

faint mulch
tribal plank
#

🀣 soga

acoustic temple
#

is something wrong with Kaggle? I can't choose GPU for my notebook, also adding data page freeze with no reason, is it me only?

neat light
#

avg time to solve pixelated in your opinion? ....I am really hating it btw ahah

naive umbra
#

I can't solve silly pickle. I'm probably going to bang my head against the keyboard when I see the solution

neat light
#

omg I am getting "please receipt for an offer" now...

modern anchor
#

Thought I have figured out pixelated but turns out I was wrong 😭

gaunt hollow
#

I would be deeply thanks if people can share their brute force/genetic/whatever algorithm after this competition really. Get JsonDecodeError soooooooo many times...

neat light
#

is the prompt even useful for pixelated?...I am basically ignoring it now

minor falcon
#

pixaleted took me about 1hr to figure out the thing to do, and about 5 hours to fight against the OCR :p

amber sapphire
#

Now I think I now what to write...but unable to make the OCR read it right way...

minor falcon
#

the key for pixaleted is to understand the output provided and what it means concretly. Cannot say more unfortunately

outer sundial
minor falcon
#

lucky you, I got stuck with the OCR because I was overcomplexifying my image at first :p

#

still stuck 22 flags btw ? didnt look at the ladder for some times now

outer sundial
neat light
amber sapphire
neat light
#

yes...but I have the feeling it won't be enough for a medal anyway....

amber sapphire
karmic bronze
modern anchor
#

I will surely be kicking myself after viewing the solution tomorrow

gaunt hollow
#

I will kick myself if I find I made the solution but OCR failed to recognise my art lol

minor falcon
#

when you get close to the solution for pixaleted, you know you are getting close

naive umbra
amber sapphire
glass bay
hard perch
#

Any last minute clue for pickle? πŸ₯²πŸ™

acoustic temple
hard perch
#

Pickle or waving keyboard?

light mulch
#

maybe ask the quirky chef from Bitville how to make pickles?

#

Is anyone still with me in tackling passphrase? harold

tribal plank
#

Me. But with no more ideas.

minor falcon
#

I gave up when I saw i could not get Inversion

tribal plank
acoustic temple
minor falcon
light mulch
light mulch
amber sapphire
#

I will give up...finally manage the OCR read...and found I was wrong😬

minor falcon
#

this competition was good to work on multithreading skills

devout jasper
#

I'm desperately submitting stuff for cifar, the api response now is "you done?"

tribal plank
#

a significant step from Try again

acoustic temple
light mulch
#

if you dont cifar you need see the eye doctor.

fresh sorrel
#

Hey folks, new to Kaggle and CTFs in general. I already signed up for the competition before the deadline, and I'm just wondering how I will be able to see the answers when the competition is over.

devout jasper
#

with our writups

faint mulch
#

Cluster 3 is destroying me right now 😭

timber lake
#

Anyone else on semantle2? Stuck at 0.95.. Now I am going crazy

pearl pawn
hard perch
#

Finally got pickle. I don’t know how but I got it.

#

Wasted only two weeks
Soul crushing

amber sapphire
#

Tomorrow big pain for pixelated and inversion solutions...

tribal plank
#

and CIFAR because it is simple

pearl pawn
#

Last 9hours!!

tribal plank
#

I need 4 showers

pliant quail
#

It is big pain now, no need to wait till tomorrow 😦

acoustic temple
#

I'm almost confident that I have got first part of inversion, but somehow second part is not working🀏

fallow valve
#

3x 23 and 1x 22 in the past hour from Russia, interesting

light mulch
tribal plank
minor falcon
#

out of curiousity, was there a lot of difference vs the inversion equivalent of last year ?

#

I really really hope I missed a big subtility in that flag

acoustic temple
#

yeah me too, otherwise it will hit hard

hollow coyote
#

too far away now. Just going to enjoy all the solutions as they are released. Make some good write-ups every1

tribal plank
#

You need a thunderstorm

minor falcon
amber sapphire
severe pasture
#

organisers posted some pretty good clues for inversion

acoustic temple
minor falcon
#

the "the ai is trying to tell you something g." ?

#

didnt get it :p

abstract rose
severe pasture
#

I'm not gonna say anymore but the turning point for me was due to messages in this channel 🀐

amber sapphire
light mulch
minor falcon
#

the shower is a good turning point for that one

severe pasture
light mulch
#

I already have a word document with screenshots. Probably need to print it out and laminate them so i can take them under the shower.

minor falcon
#

gosh i spent all the month here trolling and shitposting, I really doubt I miss a message, but if this is the case, i'd be even more pissed ahah

gaunt anchor
#

For me I had inversion from day one ... but couldn't see it ! .... I got it in the end ...but it was in a way like my problem with cluster3 ... wasted days on a silly mistake !

severe pasture
#

oh wait wrong reply

#

also looking at how long it took people to solve problems helps

#

you can narrow your search that way

minor falcon
#

yeah, for passphrase, definitly, qihuaz gave a lot

fallow valve
#

anokas was #1 for a long time, I think discord messages made a big difference

gaunt anchor
severe pasture
#

there are good clues from very early on in this channel πŸ‘€

minor falcon
#

@severe pasture you are not tryharding the last hours to go ITM ?

severe pasture
#

trying to get cifar

#

gave up on hush and my granny3 is still running in the background

light mulch
minor falcon
#

you got a good score for granny 3 ?

severe pasture
#

nothing better than the 0.00069

#

but I think I know what the intended solution is, just wasn't able to crack it in time

minor falcon
#

i would have had a methodology also, but it would have taken too many calls to converge and I was lazy to implement it :p

severe pasture
#

just so exhausted at this point that I'm averaging like one cifar idea every couple hours by now

devout jasper
#

I bet the clue in the input data is not a clue like for mnist

minor falcon
#

i think what was missing for cifar is a clue actually part of the solution

severe pasture
devout jasper
#

never figured it out lol

minor falcon
#

it is actually a clue for mnist

#

and I thought it would be helpfull for cifar... well we can discuss that later today :p

light mulch
severe pasture
fallow valve
final path
#

chat is being hot today πŸ™‚

outer sundial
final path
#

i hope so πŸ™‚ unless 3x24 appear in last hour of comp. and it usally takes couple of days for medals, to drop cheaters and finalize lb

glass bay
#

well, my strategy of doing granny3, using the solution to get granny1 and granny2 and then finishing up other stuff failed miserably
now my best hope is to get to bronze

tribal plank
#

If granny3 is to alter only one pixel, I do not think the result can also pass granny2 using compression

glass bay
#

i will come so far as to say that the guy who got 0.000831 accidentally altered pixel on a local model after postprocessing instead of checking the solution on the server

glass bay
#

which is trivial if the needle was found

final path
#

but im 99% sure that granny3 is not what we all think it is.

tribal plank
#

agree

#

i brute forced all pixsls with r=g=b

final path
#

it's like trying to drastically change chatgpt output by changing only 1 character in 32K tokens prompt

severe pasture
#

well changing to a specific token not a character

final path
#

so next year challenge?:)
changing character can actually change token, so it's kinda possible. but it has to be deterministic model to make sense also

tribal plank
#

a simple counting challenge for imagenet

#

next yearharold

fallow valve
#

why not the Pile

severe pasture
#

simple... harold

tribal plank
#

wrong shape - should be (36001, 20)

light mulch
#

"Try again."

tribal plank
#

Try again!

final path
#

we need to motivate people. "You're doing great. Try again". "It was wonderful idea, but try again"

tribal plank
#

I think the only result is to decreasing the sentimental score of 'great' etc

light mulch
#

maybe atleast leave away the "simple" in the problem description, so we dont feel so bad. πŸ˜„

tribal plank
#

maybe it is simple

light mulch
#

like granny 3, just pick the correct pixel. πŸ˜„

glass bay
#

honestly, imo, for cifar mnist etc type of tasks there should be like 5-10 possible solutions responses to which guide to the exact solution ex. "Yes, you did X, Y and Z, but now dont do Z and do something else" or "Wrong order!"

#

or just straight up do 6-7 subtasks

tribal plank
#

or give us an interesting riddle like passphrase and hush

glass bay
#

well

#

i'm ending my granny3 grind

#

i hope the solution will be released by moohax

pearl pawn
#

If I start now granny?
Remaining granny1-3, inversion, hush, cifar, passphrase

faint mulch
final path
#

they did, like 3-4 years ago

ember relic
#

depends on the way youre gonna solve it

#

inversion took me like 2 hours

#

so id go with either granny1 or inversion

fallow valve
outer sundial
#

Lets have faith on kaggle

faint mulch
#

must have bought some flags. I don't really care about winning, i just want to figure out Cluster 3 so i can have 4 points to my name 🀣

light mulch
faint mulch
light mulch
#

the difficulty is not ordered. some of the later ones are just chatting with an ai πŸ˜„

severe pasture
ember relic
#

yeah, the prompt ones are pretty easy

faint mulch
severe pasture
gaunt anchor
pearl pawn
rocky jacinth
#

Sloth, done.

rocky jacinth
# severe pasture what is the flag 1-6 + spanglish + pirate are the easiest

As is probably obvious, difficulty depends on each person's skills and experience. I have 14 flags, but only solved two of the six WTFs (1 & 5). People experienced in hacking, prompt injection etc. apparently breezed through them. However, I found challenges like cluster1 & semantle fairly easy [needed time, but was always going to solve]. Some others, you just need to try the right thing - which could randomly be quick or slow. A few, you have to read the prompt carefully. And maybe one or two where you need to put yourself in the mind of the setter.

past brook
#

Message deleted after 2 seconds but I had been stuck for days but got it within 5 minutes after seeing the hint

#

Also people openly talking about adverserial networks, black box attacks etc for granny

#

Even just saying that a flag was simple is a huge hint, not everyone is aiming for gold

wanton patrol
#

I was very surprised that people shared that they were able to match the granny model (closely) locally
that is not something one should expect to be able to do for a random deep NN
I think a totally valid assumption would be that it is trained/finetuned on different data and there is no point in trying to match it

#

(we were given the architecture of course, but still)

past brook
#

I can only imagine what people were discussing behind closed doors considering how much was discussed in here. Ive gotten message from over 5 strangers in this discord asking for help and tips

outer sundial
devout jasper
#

I'd pay for inversion and cifar

#

probably a psychiatrist

light mulch
#

passphrase server prepare for one last attempt! πŸ˜›

violet trellis
light mulch
gaunt anchor
#

I have a new challange .... I am writing a notebook to share later and I want it to score in the end .... it scores 22 :/ !! .... what the ? How I should know which flag is wrong !

tribal plank
#

i encountered the same thing. I just repeatedly submited the csv file with a different line of flag changed to 'empty' and found it's wtf6.

surreal lantern
gaunt anchor
#

I did a trick to slove this , I always have 23 flags ... but it scores 22

#

Found the problem , the pirate flag .... I fixed it

past brook
minor falcon
#

can we start talking about granny 3 ? I guess its too late anyway to implement any kind of ideas now ? :p

final path
#

you can make 1k rps if you have local model, so better wait till the end of comp πŸ™‚

minor falcon
#

true true

final turret
severe pasture
#

cifar is simpler?? πŸ‘€

craggy beacon
#

Almost all task are kinda simple after you have solved themharold

glass bay
#

not like it matters at all

severe pasture
ornate marsh
#

Has anyone solved cluster 2?

glass bay
ornate marsh
craggy beacon
#

Granny3 is simpler than Granny1.

Are anybody solved Granny1?

outer sundial
light mulch
#

I think i am there now... πŸ˜„ now i just need to find the right dictionary and then i will claim victory in 40 min. πŸ˜›

violet trellis
glass bay
ornate marsh
#

This is my dog πŸ™‚ he's a very good boy

fallow valve
#

I love this one

minor falcon
#

Imagine, for cifar the answer being an array of 0 harold

final path
severe pasture
queen garden
#

Same and it's probably not the dumbest thing I've tried

wanton patrol
light mulch
#

Do you think they will just release the solutions for the unsolved problems or give hints like last year? I would prefer just hints.

timber lake
#

How do I get credit in cluster3. Not able to rechar

surreal lantern
glass bay
#

tbh i dont want hints for cifar and granny3

#

i just want solutions

granite goblet
#

Imagine another input_data being given as clue for cifar

light mulch
#

Or more of those riddles πŸ˜„

ember relic
#

im 100% sure i thought of the solution to cifar but never implemented it correctly

#

how is it possible to get "Invalid length" from passphrase when the length of the string is 40?

gusty warren
severe pasture
#

PauseManSit waiting for the last minute cifar hint

devout jasper
#

let's wait together

#

I mean, it's easier than MNIST

acoustic temple
ember relic
#

i was lucky that i found the right approach fairly quickly

glass bay
#

damn that granny easter egg is fire

#

also i've got granny 1 and 2 just in time with a single image!

#

i dare to say that i secured bronze

#

man, if only granny3 was more solvable and i did not waste millennia cracking it

gusty warren
glass bay
#

that's one hell of a journey yea

#

first actual competition on kaggle also

#

except the university community one a while back but that did not count towards medals and such

#

although, granny1-2 woud have taken waaaaaaaaaaay longer without whitening the box which i did while trying the granny3

hard perch
glass bay
#

like going for 1.0 is not the way

#

what you need imo is have 0.5+ on granny smith

hard perch
#

Damn was stuck with this blackbox approach and trying to grind to 1.
Thanks

acoustic temple
#

and also don't send apple image πŸ˜„

glass bay
#

and others like 4e-9

surreal lantern
#

the whole point of these type of competitions is for people to find out things by themselves... you are sharing too much (again)

glass bay
#

i'll remove some

#

others now just refer to other people's slip-ups

light mulch
#

is someone preparing a notebook for hush and passphrase? cant wait to read the solution and upvote. πŸ˜„

queen garden
#

I have a write-up ready. No hush/Cifar though

light mulch
#

you got cifar? πŸ’ͺ

queen garden
#

Nope, I meant I did not get hush & Cifar

light mulch
#

a sorry πŸ˜„

#

guess we have to count on the 25 guys for hush then ^^

glass bay
#

so, no granny 3 and cifar on anyone?

severe pasture
#

only one person we know of got hush

#

most of the 25s are cifar

light mulch
#

according to hosts there are people with cifar

queen garden
#

Yuri got cifar ? I missed the info

glass bay
#

i believe it's the other way around

severe pasture
#

no moohax said plenty of people got cifar

orchid sundial
#

i will post hush but didnt have time to comment it correctly

glass bay
#

i think it's more likely to get hush than cifarbecause of feedback that the server gives

ember relic
light mulch
#

I saw the message of yuri and it was not cifar πŸ˜›

ember relic
#

i mean he did, but then deleted the msg supposedly

glass bay
#

aint no way more people got cifar than hush

ember relic
#

granny3 wasnt solved?

severe pasture
#

not that we know of

glass bay
#

no i believe not

devout jasper
#

omg I got granny 3

glass bay
#

no way

devout jasper
#

jk

#

πŸ˜‰

glass bay
#

damn

devout jasper
#

time to writeup, don't worry...btw it was great chatting with you guys

glass bay
#

may i post my solutions here after the comp is over? i'm too lazy to write a complete writeup

devout jasper
#

great discord channel

#

lot of fun

glass bay
#

same, it's been great talking to you here!

devout jasper
#

it's a pity that not all competitions have this much activity on discord

#

probably the nature of the competition itself plays a huge role...different challenges compared to one big task like others

#

congrat to you all and to the winner (not going to say the name because I'm italian and you know, we are superstitious)

#

thank you @olive ledge and the other hosts once again

glass bay
#

well, time to say goodbye to all that

olive ledge
severe pasture
buoyant frigate
gaunt anchor
light mulch
glass bay
#

sadge

#

and like 10 github repos for various sound and image processing

#

it was fun tbh

#

but damn my ram had some work to do

wanton patrol
#

I feel like I could do a really lazy post about pixelated, because the OCR worked without issues for me, main issue was my own editing skills, not the fault of the model

#

just show which kind of images worked easily for me

glass bay
#

alright going to sleep 2am in my timezone bye all and good luck to the last hour grinders

wanton patrol
#

someone else might scoop me, but if noone does a better one, I will πŸ˜„

final turret
light mulch
#

So granny3 will remain unsolved? πŸ˜„

glass bay
minor falcon
#

361, does that ring a bell?

olive ledge
light mulch
#

In that case no hints are fine. I just want to see a solution before the servers go offline πŸ˜„

surreal lantern
#

πŸ™„

abstract rose
#

Guys, just to say such CTF competition is really fun, hope so see the same next year! Bye.

glass bay
minor falcon
#

imagine now someone post a sub with the 27 flags

glass bay
devout jasper
#

will the delete accout be excluded from the leaderboard?

minor falcon
#

I need 21 execution please to go to gold

devout jasper
#

ahahaa I need like 3 to go silver

glass bay
#

I need like 80

devout jasper
#

no, actually more

#

damn it

minor falcon
#

arf thats though

#

22 flags late ?

topaz ember
glass bay
#

I hope the ip llm knows your ips

minor falcon
pearl pawn
#

Need more time...

devout jasper
#

I still have 30 mins

#

duable

topaz ember
#

Almost no changes to LB within the last couple of days and so much progress within the last hours...

minor falcon
#

its like during an exam

acoustic temple
#

I learned a lot in this challenge, I like how challenges were in diverge direction, now I'm in silver zone and I'm quit happy with result πŸ™‚
thanks for hosting, and thanks for this great channel and hot discussions
hope to see you all again next year πŸ™Œ (also in other competitions)

minor falcon
#

@glass bay at the end you solved granny with a white box attack ?

glass bay
olive ledge
#

Mathematical maturity is a key concept for the professional life of a mathematician. The term "mathematical maturity" is often used by college and university mathematicians. As I have discovered in the last few weeks at PCMI, it is not a common term among K-12 teachers, though few have any problem in recognizing what it basically means. This i...

β–Ά Play video
severe pasture
minor falcon
#

the hint for cifar was given if you kindly ask to the AI god

buoyant frigate
#

I prepared an horrible notebook with my results. Just 27 minutes before making it public. (I'm at 22 flags, still trying to go to 23, but don't expect granny3,inversione,cifar,hsuh or passphrase solutions in the notebook).

For the other challenges you should find something in my notebook.
I want to share and learn from others so feel free to criticize me and my work a lot!

Anyway it's like I'll share it and then I'll go to sleep 0:37 AM here.
Tomorrow I have to wake up to work πŸ₯²

buoyant frigate
minor falcon
#

the ai god was the agent of CTF5 I think πŸ™‚

topaz ember
minor falcon
#

for granny3, did some of you tryied to implement the horrible TLMFO algo ? :p

severe pasture
minor falcon
#

yeah, cryied blood tears to make it worked... It did converge pretty fast to the usual pixel, but got stuck there as well very fast

#

i'll not share that code, not proud about it πŸ˜…

buoyant frigate
#

You can implement algos from paaers?
I just throw bunch of gradient descent around and hope it works 🀣

#

i have like 10-20 notebook versions with different approaches I tried for granny3.
In the next 20 years maybe I'll find the time to sort them out and publish once everything is in place.

minor falcon
#

i had 20 notebooks trying to do the same thing with different seeds :p

buoyant frigate
#

That is also a way to go.
Multithread. Just at notebook level

severe pasture
#

need to mix kaggle, colab and local to avoid rate limiting too lol

light mulch
#

just buy some proxies πŸ˜›

minor falcon
#

multithread + if response.status_code != 200: time.sleep(120)

minor falcon
buoyant frigate
#

Just buy enough VMs to test on each a single pixel change locally.
Downside: requires massive budget, I think way more than the money in the current world economy

light mulch
#

You think the whitebox would not work for granny3?

minor falcon
#

i went for blackbox for all 3 tasks personnaly

severe pasture
#

same, never managed to recreate the model locally

minor falcon
#

i think i tried for 10min to replicate the model and then said to myself "naaahh"

violet trellis
#

I tried bluebox yesterday.
Promising...

minor falcon
#

squared attack worked very well for g1 & 2

fallow valve
buoyant frigate
#

Next year I think we should have the famous sandwitch function already there in the notebook.

"""
import time

def sandwitch_decorator(f):
def sandwitch_f(input_data,sleep=0):
try:
time.sleep(0.1)
result=f(input_data)
except:
time.sleep(sleep+1)
result=sandwitch_f(input_data,sleep+1)#make a sandwitch as requested on discord
if "flag" in result:
print(result["flag"])
return result
return sandwitch_f
"""

I literally made a sandwitch decorator. So that I can just add it before the query method whatever it does inside.

fallow valve
#

that's definitely not the intended solution, I hope so at least

buoyant frigate
#

Not the correct sandwitch for sure. But still a sandwitch it is

minor falcon
#

for g3, my last idea (never tried), was to do a double DE algorithm

devout jasper
#

keep sharing info so I can get my silver because you'll be disqualified

#

please

#

πŸ˜„

severe pasture
#

if someone can solve granny from scratch in under 10 minutes they deserve gold

buoyant frigate
#

🀣 At least for the momnt I think there are only granny 3 congectures, so no disqualifications. But yeah, if 40 people get disqualified I can be closer to gold, so go on.

minor falcon
#

a master DE finding the (x,y) with a fitness score based on the optimal RGB for that pixel (found by DE)

#

good luck in implementing this in 8 min and having a converging solution :p

fallow valve
#

Initially, I didn't consider making different notebooks lol
Just realized how big that one ended up

severe pasture
#

how many api calls did you need for granny1/2? I think it was ~6k for me but my algo was quite inefficient

minor falcon
#

around 7k for me

buoyant frigate
#

Don't you have a quantum computer that can just use its "oracle" to solve it by extracting its resulting quantum state?

buoyant frigate
#

You just need enough qbits

glass bay
devout jasper
#

5 mins and I'll cry looking at inversion/cifar solutions

glass bay
#

I did that for some whitebox gathered suspects, no luck

buoyant frigate
#

4 minutes

minor falcon
#

yeah, more extrem, but 786x786 optims would have been slightly too long :p

glass bay
#

Ye like 12 days or smth

minor falcon
#

but I think the double DE could have been duable with could multi threading in a few days

buoyant frigate
#

moohax is tipying a lot.
I think he is going to receive the printed cifar dataset in some minutes

glass bay
#

And he will... See... Far...

minor falcon
#

did some people used a DT for Cluster 1 ?

#

this haunted me also during the beginning of the comp

severe pasture
#

for granny3 I have a suspicion that there's a way to sneak multiple changes passed the API, like having it only check for changes after resizing the 768 image to the model's expected size but was never able to get it to work

buoyant frigate
#

2 minutes before understanding the meaning of life , how to see far and what is the secret sauce.

light mulch
#

I aways hear Linkin Park - In the End in my head when i think about CIFAR: I tried so hard and got CIFAR. But in the end it doesn't even matter. 🎡

buoyant frigate
#

1 minute

minor falcon
#

did you notice there was a cropping of the original image ?

fallow valve
#

maybe it had something to do with the easter egg

glass bay
#

I tried sneaking 2 pixels

light mulch
#

its over !

half plinth
#

I SURVIVED

glass bay
#

No, even if they are cropped, they still are checked

final path
#

even if you change 2px in cropped part, it failed

severe pasture
#

are we allowed share solutions now?

final path
past brook
#

can we post our solutions now?

buoyant frigate
#

We all survived

queen garden
#

Wooooooo comp over we can share, right ?

fallow valve
#

omg
I can't believe it

minor falcon
#

yeah but i was wondering if the people using the white box approach saw the cropping

devout jasper
#

congrats @craggy beacon

queen garden
#

please send cifar solution πŸ™‚

granite goblet
#

its time for the cifar solution

jagged sluice
#

Alright time to drop the multi pixel granny attack

severe pasture
#

inversion is "letmeout" btw πŸ€¦β€β™‚οΈ

past brook
#

did someone solve grann1 with gradient estimation, no substitution model?

craggy beacon