About QOC
QOC is a living map of quantum optimal control: a place to learn the field, follow it as it evolves, and see where its open questions lie.
Why this site
Quantum optimal control is spread across physics, chemistry, applied mathematics and engineering, and across many experimental platforms. Its knowledge sits in reviews, theses and thousands of papers that use different notation. QOC collects the essentials in one place, organizes them into a map, and keeps that map up to date as new work appears.
It is written for anyone interested in the field: students starting out, researchers moving in from a neighbouring area, and practitioners who want to keep track of what is happening.
How it is built
- Plain text. Topics, groups and open problems are stored as Markdown and YAML files in a git repository. Every change is versioned.
- One shared map. A single taxonomy of the field connects topic articles, papers, research groups and open problems.
- Automation where it helps. A scheduled job fetches new papers from the arXiv API every day and classifies them into topics.
- Static and fast. The site is plain HTML, CSS and JavaScript, generated by a small Python script — no server, no tracking.
How papers are counted
The paper tracker and the papers-per-year chart on the home page use one explicit definition. Every day, an automated job asks the arXiv API for new submissions and keeps a paper if:
- it is listed in quant-ph — or in physics.chem-ph, physics.atom-ph, cond-mat.quant-gas, math.OC and its title or abstract contains the word “quantum”;
- its title or abstract contains at least one of these phrases (whole-phrase, case-insensitive):
optimal controlquantum controlcoherent controlcontrol pulsecontrol pulsespulse optimizationpulse optimisationoptimized pulseoptimized pulsesoptimal pulseoptimal pulsespulse shapingpulse engineeringpulse designGRAPEKrotovchopped random basisshortcut to adiabaticityshortcuts to adiabaticitycounterdiabaticcounter-diabaticquantum speed limitquantum speed limitscomposite pulsecomposite pulsesdynamical decouplingcontrollabilitycontrol landscapecontrol landscapesfeedback controlLyapunov controlrobust controldynamic decouplingadiabatic passageSTIRAPinverse engineeringtime-optimalpulse-levelderivative removalquantum feedbackcoherent feedbackmeasurement-based feedbackshaped pulseshaped pulsescontrol protocolcontrol protocols - it is not a known false positive — papers where “quantum control” means a quantum-controlled operation or indefinite causal order (e.g. the quantum switch) are excluded unless they also discuss optimal control or pulses.
Papers are counted by the date of their first arXiv submission; later versions do not count again. The current year is incomplete and is shown as “to date”, compared with the same calendar period of the previous year.
This is a transparent proxy, not a perfect census. It includes neighbouring topics that use the same vocabulary (dynamical decoupling, speed limits, feedback control) and misses papers that never use these words. Because the definition is fixed, trends over time are comparable. Changes to the term list are made in the open and the whole archive is then re-harvested.
How good are the counts?
The archive is checked by hand, and the results are kept in the repository so that they can be reproduced (last check: 2026-10-12, 7,441 papers).
- Completeness
- All quant-ph papers of six full months (Jun 2005, May 2012, Oct 2017, Mar 2021, Feb 2024, Sep 2026; 5,886 papers) were downloaded and the selection rule applied locally: the archive's API query missed none of the 220 matching papers. Five months, including the three with the lowest counts, were re-fetched independently: identical papers.
- Precision ≈ 78% (± 4%)
- Share of counted papers that are genuinely about quantum control: 383 papers from stratified random samples (10-20 per year, 2000-2026) plus 45 from the phrases added in October 2026, labelled by hand. No trend across years (60-93% per year, within sampling error; 82% for 2000-2009, 75% for 2010-2019 and 78% for 2020-2026).
- Recall ≈ 70%–85%
- Share of quantum-control papers that are counted: 1,884 arXiv papers (2005-2026) citing core QOC works (GRAPE, CRAB, dCRAB, Krotov, GOAT, the QOC roadmap and major reviews); 60% are in the archive, with no trend over time (55-64% per year since 2017; noisier before, when the reference set is small; it cannot cover the years before the first core work, 2005). Many citing papers are not about control: in a labelled sample of the missed ones, 30% were genuine QOC papers, which puts the true recall at roughly 70-85%.
- Context
- QOC papers rose from 1.3% of all quant-ph submissions in 2000 to 4.4% in 2014 and have stayed between 3.8% and 4.5% every year since: the field first grew faster than quantum physics as a whole, then at the same pace.
- Topics ≈ 60% precision, 82% recall
- How papers are placed on the field map: Papers are assigned up to four field-map topics by a phrase-based classifier (data/classifier.yaml). The rules were tuned on 60 hand-labelled papers and then frozen; on 120 different papers, labelled blind beforehand, 60% of the assigned topics were correct, 82% of the correct topics were found, and 89% of the papers received at least one correct topic. Most errors are topics a paper mentions without being about them. (One rule fix prompted by the test, for open-system control, raises precision to 63% on the same sample.)
Over- and under-counting are of similar size, so the totals are approximately right. More importantly, both are stable from year to year, so the trend is reliable.
Contribute
QOC is open: the content and the code live in a public GitHub repository. Corrections and additions — a missing group, a new open problem, a result we overlooked, an error in an article — are welcome as issues or pull requests. Every data file is plain YAML or Markdown, so no web-development knowledge is needed.
Every statement in the articles and open problems is written from the papers cited next to it, and every reference is generated from its DOI or arXiv record. If something is wrong or out of date, please tell us.
Licence
The text of the site — articles, open problems, resources, the field map and the group list — is licensed under CC BY 4.0: you may reuse and adapt it, also commercially, as long as you credit QOC and link to the site. The code is under the MIT licence. Titles, author lists and abstracts of arXiv papers are not covered: they belong to their authors and are shown under the terms of arXiv.