arXiv subjects

Yanbei Chen

Publications and source records attributed to Yanbei Chen.

At least 19 recordsLinked to original sources

arXiv2026

Curvature Reconstruction and Tensor Polarizations in Isospectrality-Breaking Black-Hole Perturbations

Testing general relativity with the ringdown phase of gravitational waves requires connecting modifications of black-hole perturbation dynamics to observable signatures in detector strain. In this work, we study the relation between curvature perturbations, metric master variables, and detector tensor polarizations for nonrotating black holes in dynamical Chern-Simons gravity. In this theory the axial gravitational perturbation couples to a dynamical scalar field, while the polar sector remains GR-like at the order considered. We first construct a map for the perturbed Weyl scalar $δΨ_0$, expressing its even-parity part in terms of the ZM master function and its odd-parity part in terms of the RW master function together with the scalar radial perturbation. We also derive the leading asymptotic tensor polarizations $h_+$ and $h_\times$ directly from the metric perturbations. The resulting strain is reconstructed from the radiative gravitational master functions: the ZM function in the even sector and the Cunningham-Price-Moncrief-normalized odd function in the axial sector for dynamical Chern-Simons gravity. No independent local algebraic term proportional to the scalar radial function appears in the leading tensor strain. These results separate curvature-level scalar dependence from tensor-channel scalar-frequency responses and provide the formal reconstruction basis for companion work on parity-split beating and effective ringdown frequencies.

gr-qc↗

arXiv2026

The optical force on a dielectric slab

Radiation-pressure forces on transmissive optics, including dynamical backaction, are usually computed from the momentum flux of the external fields. For a dielectric of finite thickness, this omits the time dependence of the electromagnetic momentum stored inside. We compute the force on a lossless dielectric slab of arbitrary thickness $L_s$ submerged in a dielectric fluid, proceeding from both Lorentz force on its bound charges and from the external momentum fluxes for arbitrary carrier and sideband fields. We treat the surrounding medium through a prescribed linear dielectric response and do not model the accompanying mechanical response of the fluid, including pressure and electrostriction. These two methods both recover the expected force in the limit of a thin reflector, but differ in the more general case by the rate of change of the field momentum stored in the slab. This storage term vanishes for static fields and at low frequencies reduces to $(L_s/c^2)\,dP/dt$, where $P$ is the net power through the slab. In a cavity, where $P$ depends on the slab position, the storage term produces a velocity-dependent force. The Abraham and Minkowski momentum formulations applied to the storage term are reconciled to give an unambiguous force on the dielectric in the case of an exterior vacuum environment, where there is no fluid response to model.

physics.optics↗

arXiv2026

Modeling extreme mass-ratio inspirals in astrophysical and dark-matter environments

Extreme mass-ratio inspirals offer a powerful probe of massive black holes and their environments, but interpreting their signals requires accurate models of orbital dynamics and radiation. Here we bring together advances in black-hole perturbation theory to establish a unified relativistic framework for adiabatic inspirals in perturbative astrophysical environments. Built on the modified Teukolsky formalism, the framework requires neither post-Newtonian nor slow-rotation approximations. As a first application, we calculate the leading cloud-induced corrections to gravitational-wave fluxes and conservative orbital dynamics for circular, equatorial inspirals around a rotating black hole surrounded by a synchronized ultralight scalar cloud, combining these corrections with the previously calculated scalar response. In our fiducial system, a cloud containing $10^{-4}$ of the black-hole mass produces about $-57.3$ radians of orbital dephasing over two years. The phase contributions from gravitational-wave flux corrections and background-induced orbital shifts each exceed the scalar contribution in magnitude and partially cancel, highlighting the need for coupled matter-gravity modeling. Extensible to other environments and modified gravity theories, our framework provides the relativistic inputs needed to assess detection losses and parameter biases, and to probe the nature of black-hole environments with future space-based gravitational-wave observations.

gr-qc↗

arXiv2026

White paper: 1-10 Hz matter-wave interferometer to test the spin entanglement witness for quantum gravity

In this white paper, we highlight the importance of the ($1-10~{\rm Hz}$) frequency range for laboratory tests of the quantum nature of gravity using the quantum gravity-induced entanglement of masses (QGEM) protocol. QGEM requires matter-wave interferometers with masses ($m\sim10^{-15}-10^{-14}~{\rm kg}$), brought within separations ($d\sim30-50~μ{\rm m}$), while maintaining spatial superpositions of ($1-20~μ{\rm m}$) and coherence for ($τ\sim 0.1 - 1~{\rm s}$). These requirements make low-frequency environmental noise a central experimental challenge and place QGEM in a regime closely related to the low-frequency goals of the Einstein Telescope (ET) and the Cosmic Explorer (CE). In particular, QGEM is sensitive to relative acceleration noise (RAN) and to gravity-gradient noise (GGN) generated by seismic and other environmental mass-density fluctuations. For representative parameters $m=10^{-14}~{\rm kg}$, $Δx=10~μ{\rm m}$, and $τ=1~{\rm s}$, the differential acceleration-noise amplitude spectral density must be suppressed well below the $10^{-15}~{\rm m\,s^{-2}/\sqrt{Hz}}$ level to keep acceleration-induced dephasing below the relevant experimental scale. Achieving this level of low-frequency noise suppression is therefore a key requirement for QGEM and closely parallels the seismic and gravity-gradient noise challenges that ET and CE address.

quant-ph↗

arXiv2026

Parity Isospectrality of the Kerr--Newman--de Sitter Black Hole

Vacuum Kerr perturbations in general relativity have identical even- and odd-parity quasinormal frequencies. We establish this parity isospectrality for linear gravitoelectromagnetic perturbations of subextremal Kerr--Newman--de Sitter black holes, without a slow-rotation or small-charge expansion. We reconstruct the radiative metric and gauge-potential perturbations directly from gauge-invariant master variables in radiation gauge, without an intermediate complex potential.

gr-qc↗

arXiv2026

Scaling Laws for Looped Mixture of Experts

Looped transformers and Mixture-of-Experts (MoE) offer complementary routes to efficient scaling: recurrence increases computational depth at fixed parameters, while MoE sparsity expands total capacity at fixed active compute. Yet existing scaling laws model recurrence or sparsity in isolation. In this work, we introduce Loop Scaling Laws, the first scaling law to jointly model recurrence and sparsity alongside model size and data. At its core is a bounded, sparsity-conditional recurrence mapping that characterizes the effective-parameter gain from looping and how sparsity raises this gain. The laws predict the held-out loss of looped models more accurately than prior alternatives, and recover the standard dense and MoE scaling laws as special cases. Beyond prediction, the fitted laws provide a principled foundation for designing looped MoE models under compute and memory constraints. Downstream evaluations further demonstrate the complementary benefits of the two axes: sparsity delivers ~3x active-parameter efficiency, recurrence yields ~2x total-parameter efficiency on reasoning, and joint scaling further advances the performance frontier. As a practical extension, we show these gains hold at trillion-token scale: at matched training compute, a looped MoE with law-derived recurrence matches a ~2x larger non-looped MoE on the reasoning benchmarks, while enabling test-time scaling through recurrence.

cs.LG↗

arXiv2026

Complex frequency evolution of direct waves from binary black hole mergers

While no signal originating at a black hole's event horizon can reach future null infinity, information about the horizon and its immediate vicinity can be encoded in asymptotic properties of waves emitted by matter or field perturbations falling toward a growing/forming horizon. Within a response-filtered framework, "direct wave" denotes source-sensitive plunge and remnant-formation information revealed by filtering the black-hole response. Unlike a stationary damped sinusoid with a fixed complex frequency, the direct wave follows the evolving source and has an evolving instantaneous complex frequency tied at late times to the remnant horizon angular velocity $Ω_H$ and surface gravity $κ_H$. Using rational filters, we remove quasinormal modes from numerical-relativity waveforms to study this evolution. Before numerical contamination, trajectories depend on remnant spin: the real frequency evolves toward $2Ω_H$ from above for lower spins ($χ_f\lesssim0.7$) and from below for higher spins ($χ_f\gtrsim0.7$), while the instantaneous decay rate increases, with best-resolved cases approaching the expected value of $3κ_H$. As in particle-plunge results, the horizon-controlled value is approached only at late times, as frame dragging controls near-horizon motion. For $χ_f\sim0.7$ remnants of non-precessing, comparable-mass binaries, the early-time real frequency is close to $2Ω_H$ because the binary orbital frequency transitions smoothly to the remnant horizon frequency. Finite-time deviations therefore carry information about merger/collapse dynamics rather than undermining the horizon connection. Full direct-wave evolution requires numerical-relativity calibration. We further show that approximate pole-zero pairing in the Kerr response motivates a minimal filter set that suppresses quasinormal-mode features while revealing source-trajectory information.

gr-qc↗

arXiv2026

On Stochastic Memory Backgrounds In LISA

Gravitational-wave (GW) memory is a permanent change in spacetime geometry induced by a burst of gravitational waves. It is theoretically interesting for its connection to asymptotic symmetries, but has not yet been observed. The upcoming Laser Interferometer Space Antenna (LISA) is predicted to detect memory directly in loud events from supermassive black hole (SMBH) mergers. We assess the prospects for the cumulative effect of memory contributions as a stochastic GW memory background (SGWMB). Using two SMBH population models, we recover previously obtained single-event SNR forecasts and compute the power spectrum of the unresolved background, using numerical-relativity waveforms for modeling the memory signal during each event. We find the predicted spectrum is significantly smaller than ones found by modeling the memory waveform with a Heaviside function, with expected SNRs of order $O(0.1-10)$. We further find that our models favor a highly non-Gaussian ("popcorn") spectrum, demonstrating that the background may not appear as the union of continuous, overlapping signals, but rather as a collection of intermittent bursts. In both measures, we demonstrate that any current prediction of the spectrum is heavily dependent on the choice of population model. We interpret these findings in the light of the future LISA mission and emphasize their importance for properly handling memory within the LISA global fit.

gr-qc↗

arXiv2026

What resources are needed for optimal learning of bosonic Gaussian states?

Continuous-variable systems enable key quantum technologies in computation, communication, and sensing. Bosonic Gaussian states emerge naturally in various such applications, including gravitational-wave and dark-matter detection. A fundamental question is how to characterize an unknown bosonic Gaussian state from as few samples as possible. Despite decades-long exploration, the ultimate efficiency limit remains unclear. In this work, we study the necessary and sufficient number of copies to learn an $n$-mode Gaussian state, with energy less than $E$, to $\varepsilon$ trace distance with high probability. We prove a lower bound of $Ω(n^3/\varepsilon^2)$ for Gaussian measurements, matching the best known upper bound up to doubly-log energy dependence, and $Ω(n^2/\varepsilon^2)$ for arbitrary measurements. We further show an upper bound of $\widetilde{O}(n^2/\varepsilon^2)$ given that the Gaussian state is promised to be either pure or passive. Interestingly, while Gaussian measurements suffice for nearly optimal learning of pure Gaussian states, non-Gaussian measurements are provably required for optimal learning of passive Gaussian states. Finally, focusing on learning single-mode Gaussian states via non-entangling Gaussian measurements, we provide a nearly tight bound of $\widetildeΘ(E/\varepsilon^2)$ for any non-adaptive schemes, showing adaptivity is indispensable for nearly energy-independent scaling. As a byproduct, we establish sharp bounds on the trace distance between Gaussian states in terms of the total variation distance between their Wigner distributions, and obtain a nearly tight sample complexity bound for learning the Wigner distribution of any Gaussian state to $\varepsilon$ total variation distance. Our results significantly advance quantum learning theory in the bosonic regimes and have practical impact in quantum sensing and benchmarking applications.

quant-ph↗

arXiv2026

From driven oscillations to free ringdown: a particle plunging into Kerr

We show that a ringdown waveform can admit a constant-coefficient quasinormal-mode (QNM) representation while its individual QNM-pole contributions remain driven. To establish this result, we construct a first-principles rational approximation to the Kerr Green's function from its QNM poles, physical residues, and horizon-frequency zeros. For particles plunging from the innermost stable circular orbit into Kerr black holes with spins 0.5-0.9, each QNM-pole contribution initially follows the source's instantaneous complex frequency and decouples only after the source decays faster than the corresponding mode. At late times, the source approaches a sum of damped oscillations at the third and higher horizon frequencies. The corresponding zeros in the Green's function cancel these source-frequency components in the coherent waveform, while the continuing drive generates oscillations at QNM frequencies. Thus, an apparently free QNM superposition can emerge before its individual contributions dynamically decouple from the source.

gr-qc↗

arXiv2026

Chiral symmetry and black hole isospectrality

We prove that black-hole isospectrality between linearly independent solutions follows whenever radiative metric and matter perturbations are reconstructed from complex master variables obeying a closed, complex-linear system with complex-linear boundary conditions. If a phase rotation maps one parity sector to the other, then the even- and odd-parity QNM spectra coincide, and the perturbations are parity isospectral. A broad class of chiral-aligned theories satisfies these criteria. As an application, we demonstrate that subextremal Kerr--de Sitter black holes are parity isospectral.

gr-qc↗

arXiv2026

Tomography of a Macroscopic Quantum State influenced by Classical Self-Gravity

Macroscopic optomechanical systems offer a promising testbed for distinguishing whether gravity acts as a quantum interaction or as a classical field. Schrodinger-Newton (SN) theory is the nonrelativistic limit of semi-classical gravity where quantum matter couples to classical gravity. Based on SN theory, this work investigates how classical self-gravity affects continuous quantum state tomography of a macroscopic mechanical oscillator monitored by variable-angle homodyne detection. In the Schrodinger-Newton (SN) theory, the measurement record arises from a different conditional test mass dynamics from that in quantum-gravity (QG)/standard quantum mechanics, consequently, applying the QG-optimised reconstruction map introduces an additional state-dependent contribution. We show that this contribution makes the reconstructed covariance depend on the chosen set of tomography angles and can drive the SN covariance--after QG filtering--outside the standard Gaussian-covariance domain set by the Heisenberg uncertainty principle. We quantify the resulting QG-SN distinguishability via the Hellinger distance and analyse its dependence on measurement strength and temperature. We then formulate the same issue in the broader setting of nonlinear quantum mechanics: when the system's conditional dynamics during the readout process depends on the state being inferred, the tomographic map acquires nonlinear, model-dependent corrections to the usual Radon or Gaussian reconstruction map.

quant-ph↗

arXiv2026

Optimized Quantum States for Sensing in the Presence of Loss and Phase Noise

Squeezed vacuum lets gravitational-wave detectors and other quantum sensors surpass the standard quantum limit, and is optimal in the loss-limited regime; phase noise breaks this optimality. Numerically optimizing the quantum Fisher information across the loss and phase-noise landscape, we identify non-Gaussian states that outperform any Gaussian state. These fall into three classes: Fock-like, cubic-phase-like, and states with discrete rotational symmetry. Limiting the average number of photons in the input state to $\bar{n}=5$, with $1-η= 5\%$ photon loss and 200 mrad phase noise, the non-Gaussian advantage reaches up to 2.2 dB. Furthermore, we observe that the non-Gaussian advantage can persist even when the measurement strategy is homodyne detection.

quant-ph↗

arXiv2026

Gaussian Quantum Metrology with Realistic Linear Sensors

Quantum sensing promises enhanced precision, but the usual quantum Cramer Rao bound can be too optimistic for realistic linear sensors, where squeezing, filtering, and loss reshape quantum noise. We derive the tight Holevo Cramer Rao bound and show that realistic degradation yields a hierarchy with the usual bound and homodyne readout. This hierarchy already exists in gravitational-wave detectors. We propose a hardware-efficient readout that reaches the Holevo bound without extra signal loss, increasing compact-binary detection rates by up to 25% over the present LIGO homodyne readout.

quant-ph↗

arXiv2026

Inspiral tests of general relativity and waveform geometry

The phase evolution of gravitational waves encodes critical information about the orbital dynamics of binary systems. In this work, we test the robustness of parameterized tests against unmodeled deviations from general relativity. We demonstrate that these parameterized tests are flexible and sensitive in detecting generic deviations in the waveform using the Cutler-Vallisneri bias formalism. This universality arises from examining the inherent geometry of the waveform signal and understanding how biases manifest. We show how Bayes factors are governed by the intrinsic geometry of the waveform signal manifold when parameterized tests are used to approximate generic violations of GR. We use the singular value decomposition to propose templates that are orthogonal to parameterized tests, identifying degeneracies and enhancing the detection of potential deviations. More broadly, the geometric framework developed here clarifies -- at a fundamental level -- how subtle waveform effects (including orbital eccentricity, spin precession, waveform systematics, and instrumental glitches) can mimic one another in data, and when they are intrinsically distinguishable.

gr-qc↗

arXiv2026

MobileMoE: Scaling On-Device Mixture of Experts

Mixture-of-Experts (MoE) has become the de facto architecture for hundred-billion-parameter language models, yet its advantages at sub-billion scales for on-device deployment remain largely unexplored. To close this gap, we present MobileMoE, a family of on-device MoE language models with sub-billion active parameters (0.3-0.9B active and 1.3-5.3B total) that establish a new Pareto frontier for on-device LLMs. We first formulate an on-device MoE scaling law that jointly optimizes MoE architecture under mobile memory and compute constraints, identifying an on-device sweet spot - moderate sparsity with fine-grained and shared experts - that is simultaneously memory and compute-optimal. Building on the derived architectures, we train MobileMoE with a four-stage recipe covering pre-training, mid-training, instruction fine-tuning, and quantization-aware training, all on open-source datasets. Across 14 benchmarks, MobileMoE matches or exceeds leading on-device dense LLMs with 2-4$\times$ fewer inference FLOPs, and matches or surpasses the state-of-the-art MoE OLMoE-1B-7B with up to 60% fewer parameters. To bridge the last mile to mobile deployment, we provide the first efficient MoE inference on commodity smartphones with comprehensive on-device profiling. At comparable INT4 weight memory, MobileMoE-S delivers $1.8$-$3.8\times$ faster prefill and $2.2$-$3.4\times$ faster decode than the dense baseline MobileLLM-Pro.

cs.LG↗

arXiv2026

Universal Bound for Entanglement Generation

We derive a universal condition for entanglement generation under general bilinear interactions in the presence of white thermal noise. While various protocols have been proposed to enhance the amount of generated entanglement, it remains unclear whether they can also relax the threshold for entanglement generation itself. Using a Gorini-Kossakowski-Sudarshan-Lindblad description, we analyze general multimode systems and derive a separability-preserving condition for bilinear interactions under white thermal noise. As an application to gravity-induced entanglement, we show that the gravitational interaction must dominate over thermal noise for entanglement to arise. In particular, this bound cannot be relaxed by changing the initial state or by introducing mediator systems, although such ingredients may enhance the amount of entanglement once it is generated. These results establish a general limitation on entanglement-generation protocols in thermal environments.

quant-ph↗

arXiv2026

Amplification and generation bounds of gravity-induced entanglement in pulsed optomechanical systems

We investigate gravity-induced entanglement between the output optical fields of two red-detuned pulsed optomechanical systems with their masses coupled by mutual gravitational interaction. For each individual system, the optomechanical interaction realizes a beam-splitter state swap between an incident optical pulse and its mechanical mode. Using two rectangular pulses for each system -- the first to imprint a nonclassical state on the mechanical modes and the second to read the gravitationally generated entanglement back onto the outgoing light -- we show that the amount of entanglement can be amplified by preparing the input in a squeezed or Fock state. However, the threshold for entanglement generation is set by the competition between the gravitational coupling and thermal decoherence, $g_G>2γ_m N_{\rm th}$, and cannot be lowered by any choice of input state. We prove this bound for two-mode Gaussian inputs and show that it continues to hold for Fock-state inputs. We further analyze how imperfect detection modifies the threshold and identify the entanglement-annihilating and entanglement-breaking regimes, which are set by the thermal decoherence accumulated over the interaction time, independent of the gravitational coupling.

quant-ph↗