Shawn Shan @shawnshan_ Twitter profile | Pikagi

Pikagi

Shawn Shan

@shawnshan_

1,626

Followers

283

Following

3

Media

313

Statuses

Security and Privacy of Machine Learning, advised by @ravenben and @heatherzheng @UChicagoCS . Forbes #30Under30 @TheGlazeProject

Chicago, IL

https://t.co/nYZqciqVnh

Joined November 2017

Don't wanna be here? Send us removal request.

Pinned Tweet

@shawnshan_

Shawn Shan

2 years

Hyper excited to announce our paper Glaze, which protects artists from AI mimicry. This is *far* from stopping AI once and for all, but we believe it is a necessary first step towards artist-centric protection tools while longer term (legal, regulatory) efforts take hold.

@nytimes

The New York Times

2 years

AI-generated images are winning art contests, adorning book covers and leaving human artists worried about their futures. “That data is my artwork, that’s my life. It feels like my identity,” one woman said. A new tool is trying to protect human-made art.

20

188

891

31

632

3K

Last Seen Profiles

@222pae

@ikemen_gaku

@rouge_torihiki

@popo_po27

@AquaBitcoinES

@AustinRikley

@SpexMfanaMan

@gabbywilliams15

@py9606

@19MoonWolf

@sessykj

@Vivianabillo

@KaspaAfrica

@gruberjl

@AimAdapt

@redbull1mia

@_Hoove

@Darkiora

@BriversWHL

@twinkboysheaven

@gyuarchive

@Gerardo___Rios

@ImaneOuaadil

@BRivasEscritora

@yukuhashi_lover

@mistymison

@HunnewellL26109

@jartigasm

@stw46

@TreyHeard_era55

@EXOClanApex

@SchneierY

@18UGNFT

@mujercebra

@chichiEstvz

@cbcmob

@shawnshan_

Shawn Shan

10 months

In this project, we looked at how hard it is to have text-to-image models generate wrong images for everyday terms. Turns out, you just need to insert 50 poison data to have "dog" becomes "cat", "fantasy art" becomes "oil painting". More unexpected results in the paper.

@TheGlazeProject

The Glaze Project, at UChicago

@TheGlazeProject

10 months

By now, I'm guessing most have already seen the news on our new project, Nightshade. Lots of artists sharing it, but here's the article from MIT Technology Review (thank you to the wonderful @Melissahei ), and a thread explaining its goals and design.

104

3K

8K

13

186

1K

@shawnshan_

Shawn Shan

1 year

Very excited to finally release Glaze. We took the extra time to make the app useable for artist and can run locally on personal laptops. Again, this is one of the first steps to stop invasive AI. We will ofc continually to improve the tool and continue this line of research.

@ravenben

Ben Zhao

1 year

It's a big day. Glaze, our tool for protecting artists against AI art mimicry, is now available for download/use at Glaze analyzes your art, and generates a modified version (with barely visible changes). This "cloaked" image disrupts AI mimicry process.

279

11K

23K

11

52

281

@shawnshan_

Shawn Shan

1 year

So excited Glaze is well received in the security community! Again, none of these would ever be possible without the support of our artist friends!!!

@TheGlazeProject

The Glaze Project, at UChicago

@TheGlazeProject

1 year

Just wanted to say thank you to all the wonderful artists who have helped make the Glaze project possible. We are honored to announced that Glaze received a Distinguished Paper Award at USENIX Security Symposium, and the 2023 USENIX Internet Defense Prize.

41

360

1K

3

36

274

@shawnshan_

Shawn Shan

2 years

Project website: Paper link:

Tweet card media

Glaze: Protecting Artists from Style Mimicry by Text-to-Image Models

Recent text-to-image diffusion models such as MidJourney and Stable Diffusion threaten to displace many in the professional artist community. In particular, models can learn to mimic the artistic...

4

34

242

@shawnshan_

Shawn Shan

1 year

@ZakugaMignon The exact same applies to researchers. I thought these models were the coolest thing when it came out. Till I realized the coolest are the artists, whose hard work are exploited by all these companies.

2

8

184

@shawnshan_

Shawn Shan

9 months

Honored to be chosen as a member of the 2024 Forbes 30 under 30 cohort, and to see Glaze get recognized on big stages. #TeamHuman @ForbesUnder30

Glaze on the 2024 30 Under 30 - Consumer Technology - While text-to-image AI tools were bursting into the mainstream, Chicago-based Ph.D. researchers Emily

8

16

131

@shawnshan_

Shawn Shan

10 months

This type of poisoning can also be used for protection. Protecting copyrighted content from scrapers. (we use clean-label poisoning, so we can embed poison in benign contents on the Internet).

2

3

80

@shawnshan_

Shawn Shan

10 months

@kortizart 😒😒They stole so much that paying back would break their business model.

1

0

68

@shawnshan_

Shawn Shan

10 months

@SemiPerfectDren That's a great question. Yes, it does. Except for this case, someone could upload a bunch of random sketches or pictures (not their finest art) and apply these artifacts on them. Then these AI trainers who are scraping entire Internet, will absorb these poison into their model.

2

0

65

@shawnshan_

Shawn Shan

2 years

@leos_ng Currently, we plan to release user end apps. The main reason is that this way, the users' original art pieces never leaves their machine. But if there are enough interests, we are happy to host a service.

2

1

55

@shawnshan_

Shawn Shan

10 months

Paper link:

Tweet card media

Nightshade: Prompt-Specific Poisoning Attacks on Text-to-Image...

Data poisoning attacks manipulate training data to introduce unexpected behaviors into machine learning models at training time. For text-to-image generative models with massive training datasets,...

2

4

55

@shawnshan_

Shawn Shan

1 year

I never thought about this implication while building the tool: even Glaze is broken by adaptive attacks, the act of adaptive attack itself has legal and ethical consequences.

@neilturkewitz

neil turkewitz

1 year

@ravenben @JonLamArt @shawnshan_ “It’s clear that excuses about ‘only scraping publicly available data’ are going to look even less convincing if they’re caught deploying measures to undo active protections applied by artists.” 🙏🏽 @riptari 👀 @ruchowdh @timnitGebru @mmitchell_ai @random_walker @emilymbender

1

14

76

0

5

52

@shawnshan_

Shawn Shan

2 years

Have been wanting to do this for a while now. I finally put together a full list of Test of Time papers at Top 4 security conferences. Hope this can help people (esp. junior folks like me) to appreciate and learn from these papers. 🤗

List of Test of Time Papers from Security Conferences

Here I keep track a list of papers that won Test of Time award at Top 4 security conferences. Last updated 2022/6

shawn-shan.github.io

0

13

44

@shawnshan_

Shawn Shan

10 months

Of course all of these are only possible because the incredible work all the artists put into.

1

0

33

@shawnshan_

Shawn Shan

2 years

@Dreni_Drenstar Unfortunately, protecting against stable diffusion requires us to have a deep understanding of how those models work. I have been genuinely impressed by these models, which has made me realize the damages they can cause to artists.

4

0

28

@shawnshan_

Shawn Shan

4 years

Check out this @nytimes article on our recent project, a first step to protect users against unauthorized facial recognition. Thanks to my coauthors and advisors @EmJoyWill @HuiyingLi9 @ravenben @heatherzheng . We would love to have you try our software! #Fawkes

@nytimes

The New York Times

4 years

Researchers at the University of Chicago want you to be able to post selfies without worrying that the next Clearview AI will use them to identify you

18

126

301

6

6

26

@shawnshan_

Shawn Shan

2 years

@ken57967735 Yes, we have tested on JPEG. But ofc, if you compress the images too much or add too much noise, the protection performance degrades. Then at the same time, those low-quality images are not very useful to mimicry the artist's style.

1

0

20

@shawnshan_

Shawn Shan

9 months

Glaze has grown far beyond its original team. Tons of ongoing effort from my labmates @the_stanleywu @JosephineCharli @wenxind_ @annaha_yj . Can't wait to see the future of Glaze and its related projects.

1

0

20

@shawnshan_

Shawn Shan

2 years

@_master117 Great questions. We discussed some of these limitations in the last section of our paper, if you are interested to take a look. In short, this is far from the perfect protection, but a step in this direction and hopefully motivates follow up work.

0

0

18

@shawnshan_

Shawn Shan

1 year

@ZakugaMignon Yeah. Hope with all the actions (legal, glaze, etc) we are taking and actions we will take. We can get ethically build generative AIs in not-too-far-away future.

1

0

18

@shawnshan_

Shawn Shan

10 months

@SemiPerfectDren Yup giving them incorrect ALT text would definitely mess up these models more. But we need to be careful here since people can easily give up on using ALT text and use some of those image captioning tools.

0

0

17

@shawnshan_

Shawn Shan

1 year

Many exciting updates are coming to Glaze soon!

@TheGlazeProject

The Glaze Project, at UChicago

@TheGlazeProject

1 year

Hi all. Realized today we have not updated folks on Glaze things in a while. There's so much happening on Glaze and related topics. Some are definite, and many are still evolving. So here's a thread. 1/n

8

129

569

0

1

12

@shawnshan_

Shawn Shan

10 months

@severian22 @neilturkewitz I would say most of these generative models have similar vulnerability. Here is a paper that explored similar space with LLMs.

Tweet card media

Poisoning Language Models During Instruction Tuning

Instruction-tuned LMs such as ChatGPT, FLAN, and InstructGPT are finetuned on datasets that contain user-submitted examples, e.g., FLAN aggregates numerous open-source datasets and OpenAI...

0

0

13

@shawnshan_

Shawn Shan

1 year

@kortizart @ravenben @MystWolven Haha I am enjoying myself with everything happening. All the wonderful support will keep me going. And oh yes, so many video games to catch up!!!

1

0

12

@shawnshan_

Shawn Shan

9 months

Despite @ravenben @heatherzheng keeps saying they are social media managers, all of these would not be possible without their guidance over many many years (7 to be exact).

0

0

13

@shawnshan_

Shawn Shan

1 year

@Aeyga_X Yes to both questions. We released the app fairly fast so ppl can use it asap but we have many ideas on further improvement that we will slowly build. Also incorporate any feedbacks from artist community.

0

3

11

@shawnshan_

Shawn Shan

2 years

Very excited to see concurrent work pushing back malicious use of image generation models!

@aleks_madry

Aleksander Madry

2 years

Our paper on immunizing images to diffusion model-powered malicious manipulation is out ! This approach, combined with policy incentives, aims to raise the cost of such unauthorized image editing. w/ @hadisalmanX @Alaa_Khaddaj @gpoleclerc @andrew_ilyas

1

12

47

0

1

11

@shawnshan_

Shawn Shan

2 years

@Sketch_Ferret Thank you Kim for your help!! This project would not be possible (literally) if you haven’t reached out to us after seeing our Fawkes project.

1

0

11

@shawnshan_

Shawn Shan

9 months

@zemotion Are they trying to scrape the images at scale or just a normal DDoS?

1

0

8

@shawnshan_

Shawn Shan

2 years

@Dreni_Drenstar At the same time, there are many alternative initiative to protect artists (e.g. legal). Likely more to come in the near future. We are simply offering one possibility, which has its pros and cons.

0

0

9

@shawnshan_

Shawn Shan

1 year

@antsstyle @UChicago @ravenben @heatherzheng OMG, thanks for this review and being so supportive. I have been kicking myself in the feet about those software bugs.

1

0

10

@shawnshan_

Shawn Shan

1 year

@StephenLCasper Awesome and very important work, Stephen!! I wonder whether we can take one step further to prove (probabilistically perhaps) that the generated images learns from the original work.

0

0

9

@shawnshan_

Shawn Shan

3 years

I am amazed at the extremely high quality reviews from @USENIXSecurity . Especially during this crazy time. kudos to reviewers and @inwyrd @kevinrbbutler

0

0

9

@shawnshan_

Shawn Shan

1 year

@megyoung0 Thanks everyone for the support 🥰. Glaze is obv a team effort with @jennacryan @em_wenger @RanaHanocka @heatherzheng @ravenben

2

0

8

@shawnshan_

Shawn Shan

10 months

@NecroKuma3 Yeah those won’t work. But the fact companies are desperately adding them shows they are concerned about legal consequences. If they are confident in their “fair use” arguments, why bother?

1

0

7

@shawnshan_

Shawn Shan

10 months

@Rahll Given everything happened recently, Wall E is probably a more optimistic view at this point. “In Time” is what I think it is gonna be like — very few controls all the resources and wealth, while the rest of us struggle to make a bare living.

0

0

7

@shawnshan_

Shawn Shan

2 years

@GSNotArt @RikuNKMR1 Yes, Glaze is not designed for the generic case. We do have some early ideas on targeting the generic models, but the situation is more complex (even some potential ethical concerns). We will likely work along those lines after Glaze's app release.

2

0

5

@shawnshan_

Shawn Shan

2 years

@yunixdraws We will be releasing a binary app in the coming weeks. At the point, we are not sure whether we will release the source code.

0

0

5

@shawnshan_

Shawn Shan

1 year

@Alkraas_art @TheGlazeProject Yes, we are aware this. I am not familiar with Kasperky, but I believe there should be a way to "whitelist" Glaze while keep the anti-virus running on the rest of the files.

2

0

6

@shawnshan_

Shawn Shan

2 years

@moyix @jeanqasaur @johnregehr Does it know about my future research? Might save me some time.

0

0

6

@shawnshan_

Shawn Shan

1 year

@megyoung0 @jennacryan @em_wenger @RanaHanocka @heatherzheng @ravenben And lots of support from the lovely artist community ❤️. @kortizart @NathanFowkesArt @JonLamArt @Sketch_Ferret and on and on.

0

0

6

@shawnshan_

Shawn Shan

2 years

Submit your papers to BANDS workshop @ ICLR'23.

@KaiyuanZh

Kaiyuan Zhang

2 years

We are organizing the first #ICLR2023 workshop focusing specifically on Backdoor Attacks and Defenses in Machine Learning (BANDS). The submission deadline is Jan 25, 2023. Please consider submitting your awesome work to our workshop and spread the word!

Tweet media one

Tweet media two

2

11

54

0

0

6

@shawnshan_

Shawn Shan

1 year

@evijitghosh Totally. There is a thing called plagiarism. And it has been around for awhile. 🤯

0

1

5

@shawnshan_

Shawn Shan

9 months

Along with my amazing collaborators @em_wenger @jennacryan

1

0

5

@shawnshan_

Shawn Shan

9 months

@thegautamkamath After some reflection, I realize I also act quite differently in ML confs compared to security confs. Thanks for pointing this out!

1

0

4

@shawnshan_

Shawn Shan

1 year

@TheGlazeProject @axl99 Yes that is correct. Ideally, we will implement something smarter like automatically detect multiple styles and switch targets.

0

0

4

@shawnshan_

Shawn Shan

4 years

A great start but blocking a single company does not solve the problem, if the same service can be easily replicated. .

Tweet card media

International probe launched into facial recognition firm that scrapes images from the internet

The U.K. and Australia have launched a joint probe into U.S. facial recognition firm Clearview AI, which scrapes the internet for images of people's faces.

0

0

4

@shawnshan_

Shawn Shan

1 year

@TheEliRiots @TheGlazeProject @ZakugaMignon Not linux specifically, but we are putting up a web service to host Glaze on our servers, so that it does not require any resource on your PC. Should be up at some point in the summer.

1

1

4

@shawnshan_

Shawn Shan

1 year

@apgil_art @TheGlazeProject @ZakugaMignon We are putting up a web service to host Glaze on our servers. Should be up at some point in the summer.

0

0

4

@shawnshan_

Shawn Shan

4 years

@chshibo My first first author paper was rejected 3 times before it got in. Tbh, it is a blessing in disguise for me. I learnt so much from those rejects.

0

0

4

@shawnshan_

Shawn Shan

2 years

@whyisyoung @NDSSSymposium My default responds to paper rejection is eating sushi buffet. The afterwards food coma will make everything better.

1

0

4

@shawnshan_

Shawn Shan

2 years

@ricpacca @YingchenWang96 Super cool work @ricpacca . Now I just hope my usenix talk is not scheduled at the same time as yours. If so, no one will show up to my talk haha.

0

0

4

@shawnshan_

Shawn Shan

1 year

@Rahll @palomamcclain Same. I was so exhausted from all the twitter trolls during glaze initial release and had to uninstall twitter. Only start to getting back recently. Life is a lot better after I give less crap about random ppl on the internet.

2

0

4

@shawnshan_

Shawn Shan

4 years

@usenix @USENIXSecurity I love Usenix so much and would love to help! Any volunteer opportunity for average grad students to do for free?

0

0

4

@shawnshan_

Shawn Shan

2 years

@micahgoldblum Absolutely agree. If we look at highly impactful papers N years later, they are often the simple ones.

0

0

4

@shawnshan_

Shawn Shan

10 months

@Rahll These are soooo good!! Bad translation of course but they are delicious. They shred the dry meat somehow and make into fillings. We had way too much of these during Covid lockdown.

0

0

3

@shawnshan_

Shawn Shan

3 years

Super excited to present our work on using adversarial patches to defend against website fingerprinting attacks.

Tweet card media

A Real-time Defense against Website Fingerprinting Attacks

Anonymity systems like Tor are vulnerable to Website Fingerprinting (WF) attacks, where a local passive eavesdropper infers the victim's activity. Current WF attacks based on deep learning...

@ambrademontis

Ambra

3 years

The schedule of the AISec workshop, co-located with @acm_ccs , is out! The invited talk will be held by @prateekmittal_ Website: #MLsec #MachineLearning #AI #adversarial #malware @adversarial_ML @trustworthy_ml @aivillage_dc

Tweet media one

1

20

26

1

1

4

@shawnshan_

Shawn Shan

4 years

@poolio One of those bugs that haunt me. To add more, cv2 resize, PIL image.resize, skimage.transform.resize are all slightly different, especially if the image is preprocessed.

0

1

3

@shawnshan_

Shawn Shan

1 year

@dataisgone @TheGlazeProject @ZakugaMignon Yes absolutely. We are receiving feedbacks from a diverse set of artists and will improve accordingly. But it would be a longer term process, as we are a small team.

1

0

3

@shawnshan_

Shawn Shan

10 months

@florian_tramer Then, of course, there is always the risk that some people think these tools are perfect and will solve all their problems. While inevitable, we could minimize them by better communicating limitations etc.

1

0

3

@shawnshan_

Shawn Shan

4 years

@2plus2make5 I always use the latex2rtf package then grammar check the rtf file.

0

0

3

@shawnshan_

Shawn Shan

4 years

Thank @kashhill for covering Fawkes. Also checkout @ravenben comments under this thread.

@kashhill

Kashmir Hill

4 years

Researchers have a tool to help you post photos of your loved ones online without feeling guilty about making them identifiable to face recognition systems:

4

129

230

0

0

3

@shawnshan_

Shawn Shan

2 years

@MIT_CSAIL @aleks_madry @Trevornoah @OpenAI @miramurati obv I should wait for the paper. But did you look at the transferability of these attacks? Like generate adv perturbation using SD, but use DALLE-2 for editing.

1

0

3

@shawnshan_

Shawn Shan

10 months

@florian_tramer Might a bit off topic. At individual user level, I totally agree with what you said: user has no 2nd shoot & nothing we can come up is remotely future proof. In a broader sense, however, I believe these tools give user leverages to push back and balance the power dynamic (a bit).

2

0

3

@shawnshan_

Shawn Shan

4 years

@florian_tramer @wielandbr @aleks_madry I am not sure where the incentive is for people to fix this? A fix probably won’t results in an additional publication, and I suspect people would rather work on the “next” broken defense.

2

0

3

@shawnshan_

Shawn Shan

4 years

@frankli714 @feamster That’s what I signed up for 😭

0

0

2

@shawnshan_

Shawn Shan

2 years

@msrocean @IEEESSP Congrats! This will be so useful. I rmbr digging related defenses for a long time when I just started in this.

0

0

2

@shawnshan_

Shawn Shan

4 years

@balzarot Why CCS has much more members never appear in other PCs🧐

1

0

2

@shawnshan_

Shawn Shan

4 years

@bviswana @CarmenJpu @mjav74 Congrats Bimal! Looking forward to reading the paper!

0

0

2

@shawnshan_

Shawn Shan

2 years

@moyix Dreambooth. A lot cleaner code, documentation, and need less GPU mem

1

0

2

@shawnshan_

Shawn Shan

3 years

Super excited to talk about Fawkes at ISG!

0

0

2

@shawnshan_

Shawn Shan

9 months

@KaimingCheng Very cool. Congrats on the paper!

1

0

1

@shawnshan_

Shawn Shan

10 months

@florian_tramer Sorry for the long rant, but @nickmvincent has some amazing work that does a much better argument for "data leverage".

1

0

2

@shawnshan_

Shawn Shan

2 years

@fbpierazzi @CysKCLInf @kclinformatics @satml_conf

0

0

2

@shawnshan_

Shawn Shan

3 years

@_ericzeng @franziroesner @yoshi_kohno Haven’t read the paper, but I believe many of these ads are from several large platforms? Like host Google ads on your website and Google will distribute the ads. If so, is quality of ads highly correlate with the platforms?

1

0

2

@shawnshan_

Shawn Shan

4 years

@realyangzhang LOL and you're going to treat me for dinner next time right?

Tweet media one

1

0

2

@shawnshan_

Shawn Shan

2 years

@TaliaRinger Specific to data poisoning on ML. We recently take a look at tracebacking the root cause of these attacks.

0

0

2

@shawnshan_

Shawn Shan

10 months

@Rahll @leeahdg So.. what are these issues you are referring to? Would love to learn. I am getting more and more interested in spotting AI art.

2

0

2

@shawnshan_

Shawn Shan

1 year

@antsstyle @TheGlazeProject The GPU memory seems to spike when too many analysis data are saved. We had to do the batch workflow, so that we can offload some resources once the analyzing stage is finished. I will limit the max number of images for now, till I figure out a way to limit GPU memory.

1

0

2

@shawnshan_

Shawn Shan

1 year

@Gordan_7up @TheGlazeProject @ZakugaMignon Actually, it would be very helpful for our debugging if you can let us know whether your PC can run Glaze.

1

0

2

@shawnshan_

Shawn Shan

2 years

@micahgoldblum Very excited for this. In the meantime, can you send me some existing references on this improvement?

1

0

2

@shawnshan_

Shawn Shan

10 months

@zydan_illust @TheGlazeProject Yeah, we are having some issues with CPU windows. Might take a bit of time to make it work.

1

0

2

@shawnshan_

Shawn Shan

10 months

@florian_tramer Balancing power dynamic can often leads to more fruitful progress (e.g. unionization/labor strikes lead to better labor laws). I would say we started to see some similar progress in the AI case.

1

0

2

@shawnshan_

Shawn Shan

2 years

@krismicinski @moyix They have June 7th I believe. We had a MR paper and they said June 7 is the next round ddl that we can submit to.

1

0

2

@shawnshan_

Shawn Shan

1 year

@TheGlazeProject @dead_em_ I suspect it is due to a corrupted resource file. Could you try deleting the Glaze resource folder? The folder is at C:\\USERNAME\.glaze. Once you delete that, reopen Glaze and it will redownload the resources. Hopefully this fixes the problem.

0

0

2

@shawnshan_

Shawn Shan

1 year

@Gordan_7up @TheGlazeProject @ZakugaMignon We are working on a few short term ideas to reduce the artifacts. Likely some updates in < a month.

1

0

2

@shawnshan_

Shawn Shan

9 months

@zemotion @ravenben @heatherzheng @jennacryan @em_wenger Wait, there is a u30 event? 🙀

1

0

1

@shawnshan_

Shawn Shan

1 year

@antsstyle @TheGlazeProject Can you send me an example character that crash the app? I tested some Japanese characters, it seems to work ok.

1

0

2

@shawnshan_

Shawn Shan

3 years

@NicolasPapernot @SloanFoundation Congratulations! Well deserved!

1

0

2

@shawnshan_

Shawn Shan

4 years

@kennyog Use IEEE template instead

0

0

2

@shawnshan_

Shawn Shan

1 year

@TheGlazeProject @antsstyle Thank you!! I will try to make these changes today.

0

0

2

@shawnshan_

Shawn Shan

1 year

@TheFutman27 @TheGlazeProject @ZakugaMignon Yeah we tried. Turns out pytorch (one of the Python library we use) has a bug with M1 chips. So we weren't able to leverage yM1. Hope we can address this in the future.

0

0

2

@shawnshan_

Shawn Shan

3 years

I had very positive experience as external reviewer at PETS. Apply for next year in the link below.

@mmazurek_

mmazurek @ bsky.social

3 years

Nominate someone you know (including yourself!) as a reviewer for papers or artifacts for PETS 2022.

4

15

16

0

0

2

@shawnshan_

Shawn Shan

1 year

@midiankai @ZakugaMignon @TheGlazeProject In the meantime, I suspect the issue is with your image file. Would you mind trying to glaze a few random images you find online? See whether they also have the same problem.

1

0

2

@shawnshan_

Shawn Shan

2 years

@biggiobattista @florian_tramer I agree. I am a big fan of defense papers that acknowledge existing breaks, and discuss things learned and potential directions.

0

0

2

@shawnshan_

Shawn Shan

2 years

@NightsongWS @ravenben @kortizart @jennacryan @em_wenger @heatherzheng @RanaHanocka @apocalynds @NathanFowkesArt We tested extensively for compression/resizing/blurring/de-blurring. Not sure about Blender though. I will take a look at it after we finish the current app dev.

0

0

2

@shawnshan_

Shawn Shan

1 year

@MakanAnsariCG @NathanFowkesArt Yes we will. We may even have something stronger down the line 🤫

2

0

2

@shawnshan_

Shawn Shan

1 year

@kortizart @ravenben @MystWolven Wait we have Octopath 2 already?? I was still playing 1. Really need some catch here lol

2

0

2