Victor V. Albert

Before this talk becomes obsolete: some thoughts on where we're headed. online.kitp.ucsb.edu/online/aiqma...

Since people are putting papers there now, @github.com repos should have CITE.md file listing a standardized bibliography entry. That way, we can automate citing repos by their urls.

v2 adds over 100 codes whose twofold transversals generate the logical Clifford group, including a [[77,34,5]] generalized Golay code. Also AxiomaticAI has independently Lean verified the decomposition of arbitrary twofold gates into Z-diagonal, X-diagonal, and CNOT networks.

imagine a hunter-gatherer who’s spent his entire life learning to survive deep in an unforgiving rainforest, then a giant resort hotel springs up right next to him with a helipad and heated pools and AirBnBs, and without missing a beat scottaaronson.blog?p=10169

2026-2027 UMD postdoc application thread. First up is the Hartree QI fellowship. umd.wd1.myworkdayjobs.com/en-US/UMCP/j... @umdscience.bsky.social @umdphysics.bsky.social

“A quantum code for which there existed a universal set of encoded gates each transversal in isolation would be extremely useful.” -- Eastin & Knill arxiv.org/abs/2610.06730

Radchenko has posted an un-authored paper that claims to prove the Stark conjecture and settle the SIC-POVM problem. www.danrad.net/papers/stark...

hits especially hard these days.

"We are witnessing a general threat to intellectual work, with misalignment between the outcome of the use of AI and its initial purpose." mathandai.org

We quantified how much closer OpenAI's improvement over MRRW brings us to GV lower bound: at most 2%. Gain over MRRW: at most ~1%.

"It's tough to make predictions, especially about the future." Giving a talk tomorrow. What do you think? A artificial scarcity, money problems, compute constraints → higher prices, restrictive APIs B future models cheaper, current models already good enough, tasks routed efficiently → lower prices

We're at a strange, specific moment. Takes range from "publications are dead by fall" to "I tried LLMs in November, they were bad." If you have an LLM-assisted paper, take full responsibility as usual. Do everything: arXiv, GitHub, website, and integrate with databases if appropriate.

After extensive experiments with AI agents, the [72,36,16] extremal code remains elusive—and possibly more intractable than we expected. This isn’t your average Erdős puzzle; it’s full-stack math + compute. We’re opening the search to the community: take a shot. valbert4.github.io/selfdual_site/

Great overview talk by Tobias Osborne for anyone trying to be aware of the LLM craze. Corroborates that we should raise standards for scientific results and that, for appropriate math, formalization should be a requirement to publish. www.youtube.com/watch?v=UtJU...

Five years of work in the making: manual marathon followed by an AI-assisted verification sprint. Thank you @phfaist.bsky.social for being a steadfast and attentive architect! Suggestions on where to go from here very welcome.

Vibe-classified all subgroups of the 4-qubit symplectic group Sp(8,2), up to conjugacy. 657007 classes vs 1369 for 3-qbit, 56 for 2. AFAIK, this is the simple group with the largest classified subgroups [cf arXiv:1306.4820]. Would not be possible without agents. github.com/valbert4/sp8

Excited to share our new work led by fantastic undergraduate research student Cassie Hopkin. We show that many translation invariant qLDPC codes can be viewed as compactifications of a single parent fracton code that is local in a higher dimension. arxiv.org/abs/2605.19298 @vva.bsky.social

Prompts are becoming a new currency. Prompt libraries are popping up. I find myself asking others for their prompts. App installs can now be prompted (cf @karpathy.bsky.social's talk). Hearing discussions on giving agents a skill file instead of writing a scientific paper. Any others?

"we found that Mythos Preview is capable of identifying and then exploiting zero-day vulnerabilities in every major operating system and every major web browser" red.anthropic.com/2026/mythos-...

The forward-looking Oratomic time-efficient architecture claims to "potentially enable runtimes of 10 days for ECC–256 with ≈26,000 qubits, and 97 days for RSA–2048 with ≈102,000." This is based on stitching together several gadgets and plausibly assuming they can work together on an atom platform.

"our most time-efficient architectures can potentially enable runtimes of 10 days for ECC–256 with ≈ 26,000 qubits, and 97 days for RSA–2048 with ≈ 102,000 qubits" arxiv.org/pdf/2603.28627

Google Quantum AI now a two-platform company. blog.google/innovation-a...

March Meeting 2026

It's official: @preskill.bsky.social now at Oratomic, a new Rydberg atom startup led by Dolev Bluvstein and joined by @roberthuang.bsky.social. Exciting times!

Exciting, scary, and thoughtful session on the future of AI at #apsmarch and possible breakdown of peer review. Matthew D. Schwartz: "AI will keep getting better and better at physics. We will not."

Exciting, scary, and thoughtful session on the future of AI at #apsmarch and possible breakdown of peer review. Matthew D. Schwartz: "AI will keep getting better and better at physics. We will not."

Exciting, scary, and thoughtful session on the future of AI at #apsmarch and possible breakdown of peer review. Matthew D. Schwartz: "AI will keep getting better and better at physics. We will not."

Lots of good pointers on how to go a good project, by Nicholas Carlini. nicholas.carlini.com/writing/2026...

Here are videos and slides of my 6 hour overview of QEC, slightly updated from my @qip2025.bsky.social tutorial. Thanks goes to Kishor Bharti and Atul Arora for letting me give this at the Quantum Meets! www.youtube.com/playlist?lis... zenodo.org/records/1874...

"the progress of civilization is ultimately grounded in people who believe in their own moral imagination, while retaining humility. We're going to need many such people to help develop ideas and institutions to navigate any possible posthuman transition" michaelnotebook.com/whichfuture/...

New many-body postdoc offering at @UMDscience; come join us! umd.wd1.myworkdayjobs.com/UMCP/job/Uni...

Nice account of someone who didn't think it was all about the CV. www.ams.org/journals/not...

Encouraging news. www.nytimes.com/2026/01/10/s...

Heard many good things about these 52-chapter intro quantum mechanics notes by Robert Littlejohn. Should be useful for those interested in quantum simulation. bohr.physics.berkeley.edu/classes/221/...

Participated in my first NSF panel: wide variety of excellent proposals, from basic-science to practical. Thank you to the other panelists and program directors for upholding high standards and competitiveness!

Recommendation letters for the Hartree are due on Dec. 8th. Please verify with your writers that your letters are uploaded!

Peter Shor: we haven't found many new quantum algorithms, in part, because we need larger quantum devices for testing our heuristics. He gives several examples of algorithms that were discovered computationally, including turbo codes, @fermilab.bsky.social quantum symposium by the SQMS center.