The Signal
Researchers show how adaptive circuits can realize quantum instruments
A mathematical framework describes how adaptive sequences can implement quantum instruments while reusing ancillary qubits. The authors establish an achievable lower bound on the tradeoff between measurement steps and ancillas, and show that some instruments that add qubits can be implemented with only the difference in qubit counts as ancillas.

A framework for adaptive implementation
A paper in Quantum describes a mathematical framework for implementing quantum instruments through adaptive sequences that reuse ancillary qubits. The study, by Soham Sau and Michal Sedlák, was published on November 10, 2026, in volume 10, page 2262.
Quantum instruments describe both the probabilities of measurement outcomes and the resulting change to a quantum state. They include measurement operations such as positive operator-valued measures (POVMs), as well as quantum channels. Implementing these operations can require ancillary qubits: temporary helpers used alongside the system being processed.
Using measurement results to guide later steps
In an adaptive quantum circuit, the classical result of a mid-circuit measurement determines which gates are applied next. The paper formalizes this approach as an adaptive sequence of instruments, or ASI, and shows that any instrument can be decomposed into such a sequence.
The paper’s summary describes how mid-circuit measurements, resets and classical feedforward allow a computation to depend on earlier measurement results. That means ancillas can be reused across successive steps, rather than all being needed at once. The approach is framed around two quantities: the number of measurement steps, N, and the number of ancillary qubits, nA, needed for the implementation.
The authors establish an achievable lower bound on the product of N and nA, and determine situations in which this tradeoff is likely to be optimal. The number of steps and the number of ancillas are therefore central parameters in evaluating an adaptive sequence. The paper’s summary characterizes the bound as depending on the complexity of the instrument.
A construction for instruments that add qubits
The paper also reports a result for instruments that transform an input of n qubits into an output of m qubits, where m is greater than n. For these instruments, the authors show that an N-step ASI can use just m−n ancillary qubits.
In this construction, those ancillas are measured again N−1 times and then used as output qubits. The result illustrates how adaptive steps can reuse helper qubits while implementing an instrument that produces a larger-qubit output. It is a mathematical construction, not a report of an experiment on a particular quantum processor.
Source
Related stories
- DLR researchers report shallower circuits for quantum data loading
- Researchers Realize a Bound State in the Continuum in Ultracold Li Collisions
- Researchers publish framework for mapping non-Gaussian entanglement
- Quantum paper bounds errors in open-system approximations
- Researchers estimate X-ray spectroscopy model at 100 logical qubits