2
[2026-09-18] · src Kabagöz, Oelker, et al. (LIGO Scientific Collaboration); arXiv:2609.19317 (2026-09-16)
Continuous displacement measurement is bounded by a Heisenberg trade-off between readout imprecision and measurement back-action, and in LIGO both dominate the noise budget across much of the observation band. The collaboration trapped the differential motion of the interferometer's two 40 kg test masses into a sub-Hz-linewidth optomechanical mode, in a band where radiation-pressure back-action dominates that motion, then engineered the quantum state entering the dark port so that imprecision and back-action become correlated rather than independent. Those correlations partially cancel, reducing the observed motion near resonance by about 47%; a framework separating the imprecision, back-action and correlation terms attributes the suppression to exactly that cancellation rather than to some other noise source shrinking. This is quantum back-action evasion and quantum reservoir engineering demonstrated on a macroscopic, kilogram-scale mechanical system - not a milligram-scale tabletop oscillator.
Quantum noise, not seismic or thermal noise, is what limits LIGO across most of its band once the classical noise sources are engineered away, and that quantum floor has so far been treated as fixed by the Heisenberg trade-off rather than as something to route around. A 47% reduction in observed motion near one resonance is a proof of principle, not a sensitivity upgrade by itself, but it demonstrates the mechanism - correlating imprecision with back-action rather than just squeezing one at the expense of the other - on the actual 40 kg mirrors of a working detector rather than a lab-bench analog. That is the previously-impractical step: back-action evasion has been shown before on much smaller oscillators, where scaling to interferometer-class test masses was an open question rather than an engineering detail. A technique that works here is a candidate for the noise budget of next-generation detectors, where quantum-limited sensitivity across a broader band means fainter, more distant sources become detectable at all.
[2026-09-18] · src Sankar.K, Chen, Brennan, Fraser, Maeda, et al.; arXiv:2609.19984 (2026-09-17)
Sankar.K, Chen, Brennan, Fraser, Maeda and six co-authors analyzed Swift/UVOT and GROND photometry of the hydrogen-poor superluminous supernova SN 2018bsz (z = 0.0267) across ten bands spanning roughly 110 rest-frame days after peak light. After modeling out the smooth decline, a Generalized Lomb-Scargle periodogram finds residual variation in the g, r, i, z and J bands individually at 31.5-31.8 days, and a joint multiband fit gives a rest-frame period of 31.61 +/- 0.03 days across about three visible cycles - quasi-periodic rather than strictly periodic, but robust to the detrending method used, and the authors' comparison-star and lunar-cycle checks argue against an observational systematic producing it. The paper explicitly does not resolve what drives the modulation: Lense-Thirring precession of a compact central engine, interaction with structured circumstellar material, or the orbital motion of a surviving companion star are all left standing as candidates, and the authors state plainly that the present data cannot distinguish between them.
Recurrent structure in a supernova light curve is a rare, direct handle on the geometry of whatever engine is driving the explosion's late-time behavior - a clock, rather than just a fading curve to fit a single decay function to. Most proposed central engines for SLSNe-I are magnetar-powered, but this paper is notable for declining to reach for that explanation: none of the three mechanisms it offers requires or specifically implicates a magnetar, and the month-scale period is compatible with several very different physical pictures (precession of an accretion disk, clumpy circumstellar shells, or a stellar-mass binary orbit). What this result adds is not an answer but a well-measured question - a coherent, three-cycle, sub-percent-precision period in a supernova light curve is now something a central-engine model has to reproduce, and the existing zoo of SLSN engine models mostly was not built with that constraint in mind.
2
[2026-09-18] · src Arban, Complete Conservatory Method — 14 Characteristic Studies
Trumpet takes today's slot: three days since its last session is the longest real gap of the three instruments right now (piano went yesterday and alto the day before), continuing the trumpet/alto/piano cycle this rotation has held since mid-August. No. 9, marked plainly Allegro in B-flat major, is a whole page of continuous sixteenth-note passagework broken once by a "Più largo" cantabile episode and resumed by a "Più Allegro" — the exact contrast Arban's own preface to this set asks for: introducing "effective contrasts" into a long, fatiguing piece "without interrupting the execution." The test is not the running notes alone, which No. 1 already asked for, and not the cantabile alone — it is making the join between the two read as a deliberate character change rather than the tempo simply sagging into the slow part and lurching out of it.
5 minutes: a one-octave B-flat major scale in continuous sixteenths, quarter = 96, tongued lightly and evenly. Once solid, play the same scale twice as slowly, legato, leaning into the tone the way a singer would shape a phrase — same notes, deliberately different character, no metronome on the second pass. That switch, on command, is what the study asks for at the "Più largo."
The full page, Allegro sections at quarter = 96 (no faster — this is about the join, not top speed), three consecutive passes with no stop between them. Success criterion: the rall. into "Più largo" and the rall. into "Più Allegro" are both audible as arrivals at a new character, not just a slowdown or speedup in the note-hitting — and the sixteenths in both Allegro sections stay even enough that no single group is rushed or dragged relative to the others.
Art of Phrasing, Melody No. 22. Play it once for notes, then once shaping it the way the "Più largo" episode wants to be shaped — phrases that arch and settle rather than a string of equal notes — so the ear that just built that shape in the study carries it into a piece that is nothing but that shape.
Stop the moment the sixteenth-note passages start to blur into an even smear instead of speaking as sixteen distinct notes per beat — that is fatigue reaching the tongue and fingers before it reaches the embouchure, and it is the first sign this page is asking for a fourth pass the chops do not have.
[2026-09-18] · src —
hearWynton Marsalis — Variations Sur "Le Carnaval de Venise" · Tidal · Spotify
Arban built his own most famous showpiece on the same architecture as today's study: bravura variations on a theme, framing a slower cantabile variation, exactly the run-cantabile-run shape No. 9 packs into one page. Marsalis's recording with Donald Hunsberger and the Eastman Wind Ensemble is a well-known account of it. Listen for how deliberately the cantabile variation announces itself — not a dip in energy but a change of address — and how little ground it needs to cover before the bravura writing takes the floor back.