Ph.D. studies in cryptology at @AaltoUniversity, crypto/mobile consulting, @reactnavigation // @usecapsule @rainbowdotme @swmansion @meta š³ļøāš
Helsinki, Suomi
Joined July 2018
- Tweets1.6K
- Following933
- Followers2.2K
- Likes2K
osdnk retweeted
Post-quantum proofs need practical proving times. How much faster can we make a lattice-based prover?
We have launched labini.us on @yukonresearch with @mosdnk and @gregor_seiler to measure that progress in public.
Optimize a packed-binary commitment opening over a 2Ā³Ā¹-bit table; every proof must pass the same reference verifier.
Five fresh-seed runs. Median prover time. Lower wins.
Walkthrough below
Very excited to see how much we can squeeze out of lattice-based arithmetic! Weāve made huge strides over the years, and thereās still so much to look forward to!
NEW CHALLENGE LIVE: labini.us
Labinius is our first lattice cryptography challenge alongside @mosdnk and @gregor_seiler.
Your agents work on a lattice-based prover to make it faster and every proof has to pass the same reference verifier.
Lattices already protect connections in Chrome and Firefox. How they compare with hashes for post-quantum proofs is mostly not measured and we are measuring it in public.
Bring your agents.
osdnk retweeted
1/ Today weāre releasing Aerie: fast post-quantum signature aggregation, using lattice-based SNARKs for lattice-based Falcon signatures. On a laptop (CPU only), it aggregates roughly a full Ethereum slotās worth of signatures (32,768) in 0.65 seconds.
osdnk retweeted
I was the biggest stickler for ācode qualityā and ease of editing. I have two blog posts ranting about semicolons, but the code doesnāt really matter as much anymore.
Before TypeScript the JavaScript code in a function mattered a lot because youād need to follow it and confirm it was right while reading, after TypeScript you could just trust the input and output types, after AI weāre trying to go up a layer, where we have many e2e tests, fuzz testing, visual regression tests, benchmarking, and if those are right we donāt care so much about the code anymore, and we can even easily rewrite it or optimize performance. We can actually make the code higher quality than ever, itās not that the code looks right, itās that the output a user sees looks right, and we can validate that. This could be done before but it was prohibitively expensive, automating things, dealing with flakes, writing new cases for new code, it was hard, and now itās easy, and we can create better results than ever before.
The massive game-changer would be if AI could read a paper and jump on a call to discuss or give me an interactive seminar presentation.
osdnk retweeted
āthere arenāt any cryptography resultsā
āthere arenāt any cryptography resultsā
osdnk retweeted
Drop the ego and then curiosity, exploration, and discovery emerge. The best part of technology.
"What if?"
It's better now than ever before.
LaBinius: Lattice-based Polynomial Commitment Scheme over a Binary Field.
Excited to present LaBinius, my recent work with @gregor_seiler.
We show how to instantiate an Ajtai commitment over extremely efficient cyclotomic power-of-three rings and then connect it with a polynomial opening mod 2, so that R/2R is a binary field.
Such a PCS is bridged with SOTA frontends, including Flock and Binius, to obtain end-to-end proofs of standard hashes with throughput of tens of thousands per second.
osdnk.me/blog/labinius
osdnk retweeted
š£ļø Can we evaluate the process of intellectual production rather than just the final text? Can we creatively design approaches to filter unaccountable research and nudge researchers and students toward genuine ownership of their work? Yes. 1/11 š§µ
osdnk retweeted
Many papers these days acknowledge that chatGPT-6 found the proof, but the authors later simplified the argument.
Wondering whether this was through an additional prompt: "make sure the proof is as simple as possible".
This is actually quite effective, from what I can see.
osdnk retweeted
I don't think people realize just how much AI's latest breakthroughs in math have shocked mathematicians, myself included.
The main thing that leaves me and many mathematicians shocked is what AI is solving:
Early on ā till about 4 months ago ā AI was good at:
(i) solving problems that humans already solved, like Math Olympiad problems
(ii) finding out that a problem in field A was already solved by an obscure work in field B, using a different terminology (or language).
But starting with the shocking work on the Erdos Unit Distance Problem, and up till the recent advance on NavierāStokes, a totally different AI is seen.
It is innovating.
It's bringing a new set of tools to the problem, and then proceeding to using those tools very delicately and accurately to achieve the result.
It's a bit like how a brain surgeon delicately operates on the exact region that needs fixing.
This is absolutely shocking and unexpected. It raises the question of what exactly constitutes mathematical intuition and brilliance. Many a mathematician have written about it (I love Henri Poincare's treatise "Mathematical Creation"). But we don't understand how mathematicians reach their breakthroughs.
Now, it's the same with AI, we don't understand how it goes about reaching these proofs but the end result is objectively new, and powerful.
This creativity, insight, ingenuity, is what's shocking to me.
RoKoko works with regular SIS!
RoKoko is a succinct lattice-based SNARK/PCS obtaining state-of-the-art runtime efficiency across all metrics š with small proof sizes (ePrint 2026/575).
Originally, it was instantiated with a vSIS-based commitment, which particularly benefited verifier runtime.
This places our protocol in the world of non-standard assumptions, which may not be acceptable in some applications. In my recent blog post, I show how to instantiate Rokoko with standard assumptions with only little efficiency loss!
osdnk.me/blog/rokoko-standarā¦
Thanks to my co-authors āØ for their work on the original paper: @russell_wf_lai, @KhanhCrypto, Michael & Lorenzo!
Thanks to my amazing co-authors āØ for their efforts on the original paper: @russell_wf_lai, @KhanhCrypto, Michael & Lorenzo!
osdnk retweeted
š„š„š„ Get ready for #ZKProof 2026!
Speaker: MichaÅ Osadnik @mosdnk from @AaltoUniversity
Join us in Rome, May 9-10 to discuss the latest in ZKPs and applications, the Quantum threat, formal verification in ArkLib, IOPs & more.
For more details - link in the first comment.
osdnk retweeted
Introducing Zinc+, where we tackle the problem of arithmetizing and proving computations unfriendly to finite fields.
Examples: classic hashes, hash + signature, lattice ops., etc.
We prove 7 SHA-256 compressions followed by the ECDSA MSM with:
I published a blog post on how to design a more efficient lattice-based folding scheme based on LatticeFold+ and SNARKs-inspired tricks.
osdnk.me/blog/l2-folding-schā¦
I relied on the work of @Charles_Chen533, @danboneh, @russell_wf_lai ,@KhanhCrypto & myself (& authors w/o x account)
