
Imagine you're in a stadium with 90,000 people clapping, and somewhere in the crowd one person is clapping in a precise rhythm — say, 261 claps per second. You can't hear individual claps over the noise. All you have is a recording of the entire stadium, and you know the rhythm exists somewhere but not the tempo, not the location, and not how the tempo is drifting over time. The brute-force approach is to try every possible combination of tempo, drift rate, and seat location until you find coherence in the noise. That's pulsar hunting in Fermi gamma-ray data, and until now it has been computationally suicidal. The committed claim: by reformulating the pulsar search as a shortest-vector problem in a high-dimensional lattice — a technique borrowed from post-quantum cryptography — Pook-Kolb, Gazith, Zackay, and collaborators achieved a speedup of roughly a million-fold over prior methods. They found PSR J1622−7204, an isolated millisecond pulsar spinning 261 times per second, using 10 hours on a standard CPU core. The previous state-of-the-art method, which crowdsourced computing via Einstein@Home, failed to detect this pulsar despite expending the equivalent of 600 CPU-years. The architecture is elegant in its indirection. Traditional gamma-ray pulsar searches tile a four-dimensional parameter space — sky position (two angles), spin frequency, and spin-down rate — with millions of trial points per dimension. Each trial requires folding the entire photon dataset at the candidate parameters and checking for periodicity. The lattice method transforms this combinatorial explosion into a geometry problem: the correct pulsar parameters correspond to the shortest vector in a lattice constructed from the data. Algorithms for the shortest vector problem (SVP) — developed by cryptographers attacking lattice-based encryption — can find this vector without exhaustively searching the grid. Each lattice reduction takes about five minutes on a single core. The ladder here is dramatic. The named baseline is the Einstein@Home distributed computing search, which represents the best existing approach to blind gamma-ray pulsar discovery. That system harnesses hundreds of thousands of volunteer computers worldwide and has found dozens of pulsars over its lifetime, but each discovery costs hundreds of CPU-years. The new method doesn't just beat this baseline — it renders it almost quaint for isolated pulsars. A single laptop core accomplishes in minutes what a global volunteer network couldn't accomplish at all for this particular pulsar. The honest caveat: the demonstration is for one isolated pulsar. Binary pulsars — where orbital motion adds more unknown parameters — remain unsolved, though the team flags this as their next target. Integrity is strong for a methods paper. The dataset is NASA's publicly available Fermi LAT archive, 15 years of continuous all-sky gamma-ray observations — anyone can download it and reproduce the analysis. The discovery of PSR J1622−7204 was confirmed by folding the full dataset at the identified parameters, yielding a signal too strong to be a statistical fluke. The paper is published in Physical Review Letters, which applies rigorous peer review. The main limitation is sample size: this is one pulsar, found with the new method, serving as proof of concept rather than a systematic survey. The milestone that matters is binary pulsars. The team explicitly names this as the next target, and it's where the real scientific payoff lives — binary millisecond pulsars are critical for pulsar timing arrays used to detect gravitational waves, and the Fermi data almost certainly contains tens to hundreds of undiscovered ones. The gap between isolated and binary searches is the additional orbital parameters (period, eccentricity, companion mass) that expand the search space by orders of magnitude. Whether the lattice method can absorb those extra dimensions without losing its speed advantage is the central open question. The successor experiment the authors didn't run is the full-sky blind survey. They demonstrated the method on a known gamma-ray source position and found one pulsar. The obvious next step is to point it at every unidentified Fermi source — there are thousands — and harvest whatever pulsars are hiding. The honest read is (c): they're saving it. This paper establishes the method in PRL; the catalog paper will follow. The combination of lattice-based search with Einstein@Home's volunteer computing infrastructure could be transformative — you no longer need 600 years per candidate, so the same volunteer network could survey the entire Fermi catalog in months rather than decades.
The analogy
You're in a stadium of 90,000 clapping people, trying to find the one person clapping at a precise hidden rhythm. Brute force means checking every seat at every possible tempo. The lattice trick is like realizing the rhythm creates a geometric pattern in the noise — find the shortest line in that pattern and you've found your clapper, without checking every seat.
Stadium clapper
Finding one precise rhythm buried in 90,000 random claps by exploiting geometric structure instead of checking every seat × every tempo
Lattice pulsar search
Finding one precise spin frequency buried in sparse gamma-ray photons by reducing the search to a shortest-vector problem instead of tiling a 4D parameter grid
Honest assessment
This is a genuine methods breakthrough applied to real data with a real discovery — not a simulation-only exercise. The million-fold speedup is extraordinary and verifiable against public data. The limitation is n=1 and isolated pulsars only; the method's value scales with the catalog survey and binary extension that haven't happened yet. If the lattice approach survives the binary parameter expansion, it will reshape how the field discovers pulsars. If it doesn't, it's still a useful tool for isolated pulsars but a narrower contribution than the headline implies.
What Is CLAIMS?
80
/ 100
BedrockNews applies a structured analytical framework called CLAIMS to academic papers and technical research releases.
How CLAIMS worksClaim
4/5
Ladder
5/5
Architecture
4/5
Integrity
3/5
Milestone
4/5
Successor
4/5
What is this paper actually committing to?
Genuine methodological breakthrough. A million-fold speedup isn't incremental — it moves blind gamma-ray pulsar search from computationally impossible to laptop-scale. Demonstrated with a real discovery, not just simulation.
The field fight
Can blind gamma-ray pulsar searches ever be computationally tractable at catalog scale, or must discovery rely on targeted radio follow-up?
Side A
Blind searches are too expensive; radio follow-up of Fermi candidates is the practical path
Held by: Most radio pulsar teams, historical practice
Side B
Algorithmic breakthroughs can make blind gamma-ray searches tractable, unlocking pulsars radio can't see
Held by: Einstein@Home, this paper
Milestone tracker
Unlocks: Pulsar timing array expansion for nanohertz gravitational wave detection; discovery of pulsars invisible to radio
Current
1
isolated pulsar
this paper (10 CPU-hrs)
Next
10-100
binary pulsars
catalog-scale Fermi survey
~1-3 years if lattice method handles orbital parameters
Verified Facts (9)
PSR J1622−7204, an isolated pulsar spinning 261 times per second, found using 10 hours of computing time on a standard processor
Previous best method failed to find the pulsar despite using enough computing power that a standard processor would take 600 years
Team used 15 years of data from NASA's Fermi Gamma-ray Space Telescope
Method developed by co-authors Dotan Gazith and Barak Zackay from the Weizmann Institute of Science in Israel
Paper published in Physical Review Letters
Each trial run of the algorithm took about five minutes on a single computer core
Lead author Daniel Pook-Kolb, Max Planck Institute for Gravitational Physics (Albert Einstein Institute)
Fermi launched in 2008 and surveys the entire sky every three hours
Lattice technique draws on algorithms developed by cryptographers trying to break encryption systems designed to resist quantum computers
Practical Implications (4)
Any group with a laptop and NASA's public Fermi data can now run pulsar searches that previously required a global distributed computing network
Thousands of unidentified Fermi gamma-ray sources become tractable targets for blind pulsar searches
Pulsar timing arrays for gravitational wave detection could gain dozens of new pulsars if the binary extension works
The lattice-to-signal-search pipeline may generalize to other sparse-photon periodicity problems beyond pulsars
More latest news
Read all