Rounding out the hat-trick, a new quantum weight reduction procedure that replaces the qubits and checks of an arbitrary quantum code with patches of surface code which are glued following the connectivity of the original code.
arxiv.org/abs/2603.04883
Thanks to my collaborators Andrew & Nouédyn.
New work out today on quantum magic state cultivation with flagged low-depth adaptive circuits that gauge logical Clifford operators.
arxiv.org/abs/2603.05429
Thanks Bence and Ben for the fun collaboration.
Here’s an artist’s (Andrew’s) depiction of the parsimonious cone (bottom) vs the old unparsimonious construction (top).
Some time ago @domwilliamson.bsky.social and I wrote a paper on decoding fracton codes. This gave us the idea to try to extend MWPM decoding to all codes. We didn't get there yet but work led by @kaavyasahay.bsky.social has made great progress showing a matching decoder for bivariate bicycle codes.
Gave a talk recently at Coogee on constructing and analysing spacetime volumes of LDPC codes & surgeries thereof. Thanks to @sunnyhe.bsky.social @domwilliamson.bsky.social and Ted Yoder for the wonderful collaboration. www.youtube.com/watch?v=Lxzw...
New work out today with a killer app for our recent fast and fault tolerant logical measurement procedure on quantum codes. Here we apply it to higher-form Clifford gates to prepare many logical magic states in parallel in constant time.
arxiv.org/abs/2601.22939
I had a great time hosting TOPO2025 this week in Sydney. thanks to all the speakers and our sponsors: sites.google.com/view/topowor...
We are offering 4 PhD scholarships in #physics and #astronomy topics @sydney.edu.au
Including:
* FRBs with @manishacaleb.bsky.social
* Quantum critical points with @domwilliamson.bsky.social
* Nanostructured materials with @boriskuhlmey.bsky.social
www.sydney.edu.au/scholarships...
⚛️ 🔭 ☄️ 🧪
Come and join us: we have 2x Postdoctoral Research Associate positions available at SMRI. Mathematicians of all areas are encouraged to apply (pure, applied, statistics). Please share with your network #MathSky
Closing December 3
usyd.wd105.myworkdayjobs.com/en-GB/USYD_E...
Some cool related works
scirate.com/arxiv/2410.1...
scirate.com/arxiv/2510.0...
scirate.com/arxiv/2510.0...
Thanks again to the quantum surgeons, Alex, Sunny, and Ted!
Finally, we introduce a general upper bound on the fault tolerance of any logical measurement scheme in terms of the spacetime region of the code that is addressed by the measurement procedure.
Our results point to a tradeoff between overhead and addressability for fault-tolerant quantum logic.
We demonstrate an equivalence between block reading and homomorphic measurement, and characterize when this equivalence can preserve fault tolerance.
In the extreme limit of measuring a single logical operator, this scheme reduces to regular lattice surgery with a time overhead that scales with the code distance.
We go on to study partial block reading We characterize the space and time overhead depending on the properties of the subcode being measured.
Here is what block reading looks like on a pair of surface codes.
Block reading is a simple class of hypergraph surgeries that measure transversal logical operators. These measure logicals across copies of a code block in parallel, in constant time and linear qubit overhead. This doesn’t require the code to be single-shot, similar to algorithmic fault tolerance.
Fault-tolerant logical measurement just got a lot faster!
In new work, we show that code surgeries based on hypergraphs, rather than graphs, allow fast and parallel fault-tolerant logical measurements with low qubit overhead (without requiring the code to be single-shot).
arxiv.org/abs/2510.14895
This was also fun to listen to.
Quantum low density parity check!
That was fun.
Very happy that I was scheduled to present first today. Talk about a hard act to follow…
Also see related work with similar results plus a bunch of other cool stuff about random input Layer Codes arxiv.org/abs/2510.06659
Applying this decoder to a family of Layer Codes reveals a strong form of partial self-correction where a growing number of encoded qubits are protected for an exponentially long time in the linear system size, up to a scale that is exponential in the inverse temperature.
& here is the correction:
The concatenated matching decoder involves rounds of minimum-weight perfect-matching on coupled surface code layers, combined with a decoder for an input Quantum Tanner Code.
How do you correct this error in a Layer Code?
In new work arxiv.org/abs/2510.09218 we introduce a concatenated matching decoder and show that Layer Codes assisted by this decoder are partially self-correcting quantum memories at finite temperature!
Layer Codes are partially self-correcting: arxiv.org/abs/2510.09218
Floquet codes fit neatly onto the heavy-hex lattice. In new work out today, we show that making full use of all the heavy-hex qubits allows us to fit two floquet codes at once. We also describe transversal gates and low-depth adaptive circuits to switch to the color code.
arxiv.org/abs/2510.05225
📢 Come be my colleague: the School of Computer Science at 🇦🇺 #USyd is hiring!
Multiple tenure-track-equivalent positions across the board (including theory!), but with a specific focus on Cybersecurity and Trustworthy Digital Systems, and ML/AI.
⏰ Dec 1
usyd.wd105.myworkdayjobs.com/en-GB/USYD_E...
Gauging versus dualities: equivalent up to constant depth quantum circuits. arxiv.org/abs/2509.22051
