Context Widows or, of GPUs, LPUs, and Goal Displacement

Scientists should read this.
artificialbureaucracy.substack.com/p/context-wi...

Is there an option to attend the workshop remotely?

Thoughtful (as always) blog post from Nicholas Carlini. "Are large language models worth it?" A nice read giving his perspective on risks of ML models.

Post: nicholas.carlini.com/writing/2025...

For people who prefer, this is the video of the talk from @colmweb.org www.youtube.com/watch?v=PngH...

Intern 2026: Fault-tolerant quantum algorithms

Our team at IBM is looking for interns! If you are interested in researching fault-tolerant quantum algorithms, please apply here: ibmglobal.avature.net/en_US/career...

New paper on arxiv: We show that a practical Markov chain for stoquastic XY Hamiltonians is fast mixing. This Markov chain is also known to empirically mix fast for Qmaxcut on bipartite graphs which if proven would resolve the complexity of Qmaxcut on bipartite graphs.

arxiv.org/abs/2509.21683

The QIP 2026 call for papers is out! QIP 2026 will be held in Riga, Latvia from January 24–30, 2026. See you there!
qip2026.lu.lv

Our *new paper* explores how flexibility in social categories like gender and race can undermine unfair norms. If we can't read an identity, we can't use it to underpin discrimination. We show even a little confusion can be powerful, and advocate identity play
philsci-archive.pitt.edu/26062/

If Charles Bukowski Wrote a Children’s Book About Emotional Regulation Sometimes you feel like a fist. Other times, you feel like the ashtray after a party no one invited you to. Both are fine. Just don’t throw the fis...

This is surprisingly beautiful. McSweeney's is like The Onion, but with more layers.

"You are not your feelings.
You are the cracked cup holding them.
Still useful.
Still capable of holding tea. Or rage.
But preferably tea."

www.mcsweeneys.net/articles/if-... @mcsweeneys.net

Accepted papers at TQC 2025. tqc-conference.org/accepted-tal.... Many interesting-looking titles!

Hello world! The quantum journal is now on BlueSky!

Does anyone know when and where TQC 2025 is happening?

Nice work! I am trying to understand this result through the lens of ETH. What does your result say that is different from what ETH implies?

Do it!

Yes, please!

Quantum Effects in Algorithms We discuss some seemingly paradoxical yet valid effects of quantum physics in information processing. Firstly, we argue that the act of ``doing nothing'' on part of an entangled quantum system is a hi...

Interesting but perhaps not exactly breaking (at least as a headline). Richard Jozsa argued, over 25 years ago, that "doing nothing" is an essential ingredient for quantum speed-ups: arxiv.org/abs/quant-ph...

👋

We prove that finding the minimum eigenvalue of the Laplacian of an Independence complex, which has information about the holes, is QMA-hard, improving from a previous result of QMA1-hardness.

2. Fermionic Independent Set is also related to another problem from homology that has gained a lot of interest among QI community: are there any holes in a topological manifold?

Constrained local Hamiltonians: quantum generalizations of Vertex Cover Recent successes in producing rigorous approximation algorithms for local Hamiltonian problems such as Quantum Max Cut have exploited connections to unconstrained classical discrete optimization probl...

1. We tried to generalize Independent Set using qubits in a previous (arxiv.org/abs/2409.04433) and turned out that it can't be harder than StoqMA. We really need the intrinsic sign problem of fermions to get a QMA-hard generalization of the Independent Set problem.

Check out my recent paper on Fermionic generalization of the Independent Set problem. I define the problem and prove that it is QMA-hard. This problem is interesting for multiple reasons: