Background: two families of multiphoton entanglement
Multiphoton entanglement has two major families. GHZ states are an equal-amplitude superposition of all-zeros and all-ones; for three photons, the two terms carry equal weight. W states are an equal-amplitude superposition of the three arrangements containing exactly one 1 (100, 010, 001), each with one-third weight.
The engineering difference is stark: remove one photon from a GHZ state and the entanglement collapses; remove one from a W state and the remaining two photons stay entangled. For building quantum networks, the loss-tolerant family is more useful.
The exponential bill of quantum tomography
Given a batch of entangled photons, you must first determine which state it is. The standard method is quantum tomography: reconstructing the quantum state from many measurements. The problem is that the required data grows exponentially with photon count—add a few photons and the measurement burden multiplies several times over.
An alternative is an entangled measurement: a single measurement that directly classifies the state, without accumulating data for reconstruction. An entangled measurement for GHZ states was proposed and experimentally realized 25 years ago. W states had no counterpart—until now.
What Kyoto University and Hiroshima University did
The team is from Kyoto University and Hiroshima University, and the paper, *Entangled measurement for W states*, was published in Science Advances by Geobae Park, Holger F. Hofmann, Ryo Okamoto, and corresponding author Shigeki Takeuchi. Kyoto University's announcement is dated September 29, 2026.
Takeuchi stated that twenty-five years after the original proposal of entangled measurement for GHZ states, the W-state counterpart has finally been obtained, along with a real experimental demonstration on three-photon W states.
The handle: cyclic shift symmetry
W states possess a mathematical property called cyclic shift symmetry: shifting the photons' arrangement periodically around the cycle leaves the underlying entanglement pattern unchanged.
Building on this symmetry, the team theoretically constructed an entangled measurement based on a photonic quantum circuit. The circuit performs a quantum Fourier transform, which reshuffles quantum information so that an otherwise unreadable pattern becomes visible. In principle, the method applies to W states of arbitrary photon number.
Trade-offs on the lab bench
The actual device used three photons. The team deliberately injected different polarization states, and the system had to distinguish three different three-photon W states, each corresponding to a different class of nonclassical correlations among the three input photons.
They used a highly stable optical quantum circuit that runs continuously for long periods without active control. This is a hard requirement for such experiments—any drift in the optics ruins the data.
The team also measured the fidelity of the entangled measurement—the probability of a correct output given a pure W-state input. Kyoto University's press release confirms fidelity was measured but gives no value; the paper itself was not retrieved in this search.
| Feature | GHZ state | W state | |---|---|---| | Three-photon structure | Equal superposition of all-0 and all-1 | Equal superposition of the three single-1 permutations | | After losing one photon | Entanglement collapses | Remaining two stay entangled | | One-shot entangled measurement | Scheme existed 25 years ago | First experimental demonstration (this work) | | Scale of this experiment | Not involved | Three photons |
Where it leads
The measurement itself is not an endpoint. Its direct destinations include quantum teleportation—transferring quantum states via entanglement without moving matter—new quantum communication protocols, transfer methods for multiphoton entangled states, and measurement-based quantum computing.
The team named two things not yet done: extending the method from three photons to general larger multiphoton entangled states, and integrating the measurement into on-chip photonic quantum circuits to move it out of the optics lab.
Context: alongside other recent progress
Contemporary coverage includes another multiphoton teleportation result from China: Jietai Jing's team at East China Normal University used programmable holographic encoding with measurement-free all-optical feedforward to arrange light into a 10×10 array, simultaneously transmitting continuous-variable quantum states over 100 channels with an average fidelity of 0.60, above the 0.52 classical limit. [Secondhand report; the original paper was not verified.]
The two works point in different directions: one scales channel count, the other adds a measurement tool. Connecting the two lines, what is missing is exactly a means of determining "whether what is being transmitted across so many channels is really that family of entangled states"—the kind of tool slot that the one-shot W-state measurement fills. [Inference.]
---
Sources: Kyoto University press release via ScienceDaily (2026-09-29); Science Advances paper *Entangled measurement for W states* (authors Park, Hofmann, Okamoto, Takeuchi).
Caveats: The specific fidelity value is not given in the retrieved reports; the original paper and DOI were not directly obtained; applicability to arbitrary photon numbers is the authors' theoretical claim, while the experiment reached only three photons.