Sign in

Mark M. Wilde

@markwilde.bsky.social
722 followers 434 following 362 posts

Associate Professor at Cornell University quantum Shannon theory, quantum information theory, quantum error correction, quantum communication, quantum algorithms, quantum computational complexity theory, quantum machine learning

PostsRepliesMedia
Mark M. Wilde @markwilde.bsky.social · 02/10/2026
The new arXiv submission limit policy at blog.arxiv.org/2026/10/01/u... has an immediate impact on those submitting to the QIP conference. In fairness to all who are submitting to QIP, the QIP policy of demanding public posting of arXiv submissions in advance of Oct. 5 needs to be revisited.
blog.arxiv.org
Fair Moderation, Equitable Access, and AI: arXiv’s Updated Rate Limit Policy
arXiv, and the scientific community at large, are facing a watershed moment. Scholarly publishing is currently changing at a rapid pace, and we are seeing a…
010
Reposted by Mark M. Wilde
arXiv @arxiv.bsky.social · 01/10/2026
arXiv has updated our policy on rate limiting for all submitters. This update was made to fairly distribute moderator time & support the arXiv community of staff, volunteers, readers & authors. Please read our announcement to learn more: blog.arxiv.org/2026/10/01/updated-r…
216372
Reposted by Mark M. Wilde
Steven Strogatz @stevenstrogatz.com · 01/10/2026
A free sample of BIG MATH (my new book with Alex Townsend) is now up on Amazon. Hope you like it! read.amazon.com/sample/03165...
read.amazon.com
Read a sample: “Big Math: The Hidden Codes That Run Our World”
Steven Strogatz, Alex Townsend
0249
Reposted by Mark M. Wilde
Markus Heinrich @markusheinrich.bsky.social · 29/09/2026
Here's the 2026 edition
051
Mark M. Wilde @markwilde.bsky.social · 29/09/2026
I can help
110
Mark M. Wilde @markwilde.bsky.social · 29/09/2026
For least year's QIP, ChatGPT says that there 146 contributed regular talks. With there already being 1614 submissions to this year's QIP, if the plan is to have the same number of talks, that would amount to a 9% acceptance rate. Of course I don't know what the plan of the PC is for this year.
040
Mark M. Wilde @markwilde.bsky.social · 29/09/2026
"Submission #1614 has been registered on the Quantum Information Processing Conference (QIP 2027) site."
001
Mark M. Wilde @markwilde.bsky.social · 29/09/2026
"Submission #1595 has been registered on the Quantum Information Processing Conference (QIP 2027) site."
110
Mark M. Wilde @markwilde.bsky.social · 29/09/2026
Do you mean ChatGPT 6 Pro Astra?
210
Mark M. Wilde @markwilde.bsky.social · 29/09/2026
ok, that's certainly interesting. I haven't yet witnessed it make a mistake and I've been checking everything line by line
110
Mark M. Wilde @markwilde.bsky.social · 28/09/2026
We just submitted something to QIP and got submission #1364
000
Mark M. Wilde @markwilde.bsky.social · 27/09/2026
Yes definitely. There will be too many this week
110
Mark M. Wilde @markwilde.bsky.social · 27/09/2026
There will be a tsunami this week on @arxiv-quant-ph.bsky.social due to the impending QIP deadline, mandatory public release of submissions, & a GPT-enabled research community. Authors are recommended to ensure their papers are online by Sept. 30 so that the submissions will be public before Oct. 5
1100
Mark M. Wilde @markwilde.bsky.social · 26/09/2026
You don't agree? ChatGPT has declared that all spaces are finite dimensional. See attached :)
110
Mark M. Wilde @markwilde.bsky.social · 26/09/2026
I didn't even notice until now that we just had four different papers appear simultaneously on the strong converse for quantum channel discrimination in the parallel setting: arxiv.org/abs/2609.27196 arxiv.org/abs/2609.28635 arxiv.org/abs/2609.29885 arxiv.org/abs/2609.30268
arxiv.org
Quantum Channel Stein's Lemma with an Exponential Strong Converse
We establish the quantum channel Stein lemma at every fixed type-I error tolerance in $(0,1)$ and prove an exponential strong converse under arbitrary adaptive strategies. For any finite-dimensional c...
070
Mark M. Wilde @markwilde.bsky.social · 25/09/2026
I always considered determining the classical capacity of the amplitude damping channel to be a major open problem in quantum Shannon theory. There is now a paper released that claims a solution: arxiv.org/abs/2609.28592
arxiv.org
Classical Capacity and Entanglement Cost of the Amplitude Damping Channel
Determining a noisy quantum channel's classical capacity and entanglement cost generally requires regularization over many channel uses. We remove both regularizations for every qubit-to-qubit channel...
160
Mark M. Wilde @markwilde.bsky.social · 18/09/2026
intended to encourage high-quality submissions & promote clear & effective communication of research. The policies in the call for papers include mandatory public posting & mandatory video submission. Detailed instructions for these 2 requirements will be released closer to the submission deadline.
051
Mark M. Wilde @markwilde.bsky.social · 18/09/2026
Glad to see this: Submission guidelines for STOC 2027: acm-stoc.org/stoc2027/ Policy experiments for STOC 2027 In light of rapid advances in generative AI and their impact on research and scientific communication, STOC 2027 is experimenting with several new policies
acm-stoc.org
STOC 2027 - 59th ACM Symposium on Theory of Computing
160
Reposted by Mark M. Wilde
Clément Canonne @ccanonne.github.io · 15/09/2026
Lots of fantastic opportunities to do an 8-week (paid) research internship with my colleagues in the School of CS at #USyd! The projects I am proposing are CS2026-27/10 Pseudo deterministic Quantum Algorithms CS2026-27/18 Feasibility of Quantum One Time Programs but the whole list is worth is read!
0142
Reposted by Mark M. Wilde
Lance Fortnow @lance.fortnow.com · 16/09/2026
I'm a fan of publishing everything. It's not like we'll run out of space on the Internet. The good stuff will bubble up through social media and AI-powered searching.
041
Mark M. Wilde @markwilde.bsky.social · 14/09/2026
I know that one well!
010
Reposted by Mark M. Wilde
Terence Tao @teorth.bsky.social · 11/09/2026
A group of 25 Fields Medalists, including myself, have made a joint declaration on Math and AI: mathandai.org . We welcome additional signatories. See also this article in the Economist announcing the declaration: www.economist.com/science-and-...
mathandai.org
Declaration — Math and AI
Read the declaration and add your name.
422055930
Mark M. Wilde @markwilde.bsky.social · 10/09/2026
Nice idea in principle but will likely be heavily biased against non-native English speakers. So it doesn't seem fair.
020
Mark M. Wilde @markwilde.bsky.social · 10/09/2026
I used to watch my health to make sure my mind would stay sharp for doing research. Now, with AI & LLMs, I can be completely drunk and totally high all the time, just supervising the research process by typing a few words and clicking some buttons while sipping my beers and smoking my doobies. Nice!
0120
Mark M. Wilde @markwilde.bsky.social · 09/09/2026
Congrats! remarkable accomplishments
000
Mark M. Wilde @markwilde.bsky.social · 08/09/2026
Graeme Smith @graemesmith.bsky.social gave a QIP 2010 tutorial called Regularization and Its Discontents :)
140
Mark M. Wilde @markwilde.bsky.social · 04/09/2026
Gilles Brassard speaking at the dinner banquet of TQC 2026: "Work on research problems that you enjoy"
0130
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
We develop the theory further and apply it to quantum networks, tensorization and strong data-processing inequalities, mixing times, quantum memories, and quantum machine learning.
010
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
A particularly satisfying result is that, for q. relative entropy, our conditional contraction coefficient is exactly equal to the corresponding contraction coeff. for conditional mutual information. This establishes a direct bridge between divergence contraction & decay of quantum correlations.
110
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
This gives a nontrivial way to quantify information loss in fully quantum settings without requiring the reference systems to have identical states.
120
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
In this paper, we introduce conditional contraction coefficients, which account for this quantum side information by subtracting the distinguishability already present in the reference system.
110
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
However, the usual framework becomes problematic when the input is correlated with another quantum system. The reference system may itself contain distinguishability that the channel never touched.
110
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
Contraction coefficients provide a precise way to answer this question: they quantify how much the distinguishability between two states decreases when those states pass through a quantum channel.
110
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
New paper: Conditional contraction coefficients and their applications to quantum networks arxiv.org/abs/2608.27171 Joint work with @christophhirche.bsky.social , Ian George, and Theshani Nuradha Motivating question: How much information does a noisy quantum channel destroy?
arxiv.org
Conditional contraction coefficients and their applications to quantum networks
Contraction coefficients quantify the loss of distinguishability induced by a channel and provide a strong form of the data-processing inequality. While standard contraction coefficients ignore auxili...
1153
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
Maybe conference papers will find a new role over journal papers, in which the conference presentation would play a role in deciding whether a work should be accepted or rejected
010
Mark M. Wilde @markwilde.bsky.social · 29/08/2026
is this in reference to a particular paper or the whole slew of papers coming out now?
200
Reposted by Mark M. Wilde
Dan Garisto @dangaristo.bsky.social · 20/08/2026
Scoop: For nearly a year, NSF has struggled to make new grants. Now we know that it will not catch up. In fact, NSF will award just 6,100 new grants—a 30% y/o/y decline, itself down from a previous avg. of 11,000+. In fact, the last time NSF awarded so few grants was in the early '80s.
nature.com
Exclusive: NSF set to issue lowest number of new grants in four decades
The US science-funding agency is withholding $1 billion of its budget so that the money can go to a special White House project.
14753471
Reposted by Mark M. Wilde
Lance Fortnow @lance.fortnow.com · 19/08/2026
Back to our regularly scheduled programming. Has any noticed that AI is getting much better at math?
blog.computationalcomplexity.org
Centaur Math
In the past, new PhD students would ask how they could succeed when they had to compete with the likes of say, Richard Karp or Avi Wigderson...
0102
Mark M. Wilde @markwilde.bsky.social · 20/08/2026
was it good? I have not been there yet
110
Mark M. Wilde @markwilde.bsky.social · 17/08/2026
I would have later bragged about that person for being a hero for having solved it
051
Mark M. Wilde @markwilde.bsky.social · 16/08/2026
what I mean is that it doesn't make sense any more for me to share questions I would have previously. Someone will just look them up in a paid LLM and then post about them. So the whole dynamic has changed. I'm not paying for LLMs myself, and so I'm not sure which questions of mine it could solve.
120
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
lauded by apocalyptic AI punk critic @vrusso.bsky.social as "truly the song of our generation" and "something to put on a syllabus for the class AI policy"
010
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
Original song "Against the 70s" available here: www.youtube.com/watch?v=UHfF...
youtube.com
Against the 70's
YouTube video by Mike Watt - Topic
130
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
The kids Against GPT Kids Against GPT The kids of today should defend themselves against GPT The kids of today should defend themselves against GPT The kids of today should defend themselves against GPT The kids of today should defend themselves against GPT
140
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
The kids of today should defend themselves against GPT The kids of today should defend themselves against GPT It's not reality, Just some conditional typicality Look what it did to Scott Consulting for the brainchild of Altman
140
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
AI sloppers, they ain't scholars, just faking it Forcing youth away from the truth, they ain't making it The kids of today should defend themselves against GPT Emptying minds with empty lies gotta make you laugh Tech-bro vendors buying true defenders of the math
141
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
The kids of today should defend themselves against GPT The kids of today should defend themselves against GPT It's not reality, Just some conditional probability He didn't work for it
140
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
Lyrics: He has a question, it sure is pestering, he cannot solve It'd take some time, boy he'd have to think, seems too involved The kids of today should defend themselves against GPT He types some words, clicks a button, and he goes to sleep The next morning, whole paper's done, oh it comes cheap
140
Mark M. Wilde @markwilde.bsky.social · 15/08/2026
Further thoughts on the GPT situation in the form of a song. Based on "Against the 70s" by Mike Watt. Audio: legacy.reverbnation.com/markwilde/so... Title: Against the GPTs
legacy.reverbnation.com
Against the GPTs | Mark Wilde
Against the GPTs by Mark Wilde, Rock music from Ithaca, NY on ReverbNation
1131
Mark M. Wilde @markwilde.bsky.social · 13/08/2026
thanks! the idea came up during a group dinner in Morrison Dining Hall :)
110