In Search of the Gravitational Wave of Fast X-Ray Transient in the Second Part of the LIGO-Virgo-KAGRA Fourth Observing Run

arXiv:2610.31227 · astro-ph.HE · Submitted 2026-10-22 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Astrophysics Radio. Generated commentary on the latest astrophysics papers.

Vera: I'm Vera.

Jocelyn: And I'm Jocelyn, here with Subrahmanyan.

Vera: We're looking at the paper: "In Search of the Gravitational Wave of Fast X-Ray Transient in the Second Part of the LIGO-Virgo-KAGRA Fourth Observing Run".

Jocelyn: Here's the gist of the paper. The study proposes a methodology to search for possible relations between gravitational waves (GWs) and Fast X-ray Transients (FXTs), utilizing the unmodeled search algorithm X-Pipeline to identify excess coherent signals in LIGO-Virgo-KAGRA (LVK) O4b observations Fast X-Ray Transients (FXTs) are characterized by abrupt X-ray emissions spanning timescales of seconds to several hours and are spatially distinct from any persistent X-ray sources or previously identified stellar objects.

Vera: FXTs represent a heterogeneous class of short-lived high-energy events, whose physical origins remain under active investigation.

Jocelyn: The observed diversity in peak luminosity, duration, spectral hardness, and host-galaxy environment suggests that multiple progenitor channels may contribute to the overall population.

Vera: Major classes of proposed astrophysical origins include outbursts from X-ray binaries (XRBs), shock breakout of core-collapse supernovae (CCSNe), off-axis gamma-ray bursts (GRBs), tidal disruption events (TDEs) of systems with a white dwarf and an intermediate-mass black hole (IMBH), binary neutron star (BNS) mergers, proto-neutron stars in supernova models, collapsars in gravitationally unstable neutrino-cooled disks, and cocoon breakout caused by jet-envelope collisions.

Paper summary: Vera: We just talked about how the paper is using X-Pipeline to look for gravitational waves correlated with Fast X-ray Transients, focusing on the O4b run data from LIGO, Virgo, and KAGRA >

Jocelyn: The core thesis of this paper is proposing a way to search for possible relations between gravitational waves and FXTs because it could serve as a discrepancy among proposed theoretical models >

Subrahmanyan: So they are looking for evidence that these two things might be linked, which is crucial if we want to solve the degeneracy in the scenarios involving jet-envelope collisions or other mechanisms that cause these X-ray bursts >

Vera: They use the Einstein Probe data, which has increased our discovery rate significantly, and they focused on sixty-four FXTs events to build their search sample >

Jocelyn: They specifically selected nineteen of those FXTs for the actual robust searches because that's where they got enough data for a solid analysis > <ref:2610.31227#pg2>

Subrahmanyan: The paper claims that this methodology is important because it gives us a way to test these different physical origins—like tidal disruption events or binary neutron star mergers—by seeing if the GW search yields anything >

Vera: They outline the setup by detailing three preliminaries they need: which detectors they use, how wide their on-source window is, and what data streams and sky positions they are looking at >

Jocelyn: Specifically for this paper, they use all three operating detectors in the O4b run: LIGO-Hanford H1, LIGO-Livingston L1, and Virgo V1 >

Subrahmanyan: And the on-source window is defined as the expected arrival time of a gravitational wave counterpart relative to the targeted electromagnetic burst from the FXT >

Vera: That's what they mean. They are trying to establish that specific temporal relationship between when we see an X-ray flash and when we expect a gravitational wave signal to arrive >

Jocelyn: In contrast, the off-source window is set as a ten thousand eight hundred-second data stream during which we expect no GW events to occur, centered on the triggered time of each event > <ref:2610.31227#pg3,data stream during which we expect no GW events to occur>

Subrahmanyan: This off-source window is set based on previous work from P. J. Sutton et al. two thousand ten about background noise and its expected characteristics in that time frame > <ref:2610.31227#pg1>

Vera: So they're using those established background estimates to define the region where they are looking for an excess signal above what we expect from just noise >

Jocelyn: The authors use X-Pipeline, which is an unmodeled search algorithm, to construct a time-frequency map and then cluster the highest intensity pixels as candidate events >

Subrahmanyan: That mapping process aims to produce a coherent set of h times and h plus according to the maximum likelihood method >

Vera: After that, they apply a glitch rejection process using a threshold function involving log10 I equals alpha (log10 E) squared plus c to filter out noise events >

Jocelyn: The goal of this whole sequence is to identify the loudest event remaining after glitch rejection, and then calculate a p-value comparing it to the background loud events >

Conclusion: Vera: So, looking at the paper "In Search of the Gravitational Wave of Fast X-Ray Transient in the Second Part of the LIGO-Virgo-KAGRA Fourth Observing Run", what we see is that they used a very specific methodology to check for GWs related to FXTs >

Jocelyn: The authors are searching for any possible gravitational wave signal, and after running their analysis on nineteen events, they found none of the FXTs had a significant signal > <ref:2610.31227#pg2>

Subrahmanyan: The implication is that based on this specific search run, we haven't found evidence linking these two phenomena yet, but the paper sets upper limits for how far away those signals could be >

Vera: They set a ninety percent confidence level exclusion distance for each source according to the upper limits of possible waveform models they tested against >

Jocelyn: The paper shows that the search method has a steady sensitivity on NSBH mergers but slightly weaker sensitivity on BNS in this specific analysis >

Subrahmanyan: The authors also highlight that future work could look at extending the on-source search window by considering other potential GW mechanisms or perhaps using a single detector to enhance their sample size >

Vera: So, they're not saying there are no relations between them, just that this particular search didn't pick up anything significant yet and they are suggesting ways to improve the search for the future >

Jocelyn: It’s about using the data we get from things like the Einstein Probe to keep pushing these ideas forward by refining our search parameters >

Po-Ya Wang, Albert K.H. Kong, Surojit Saha

Institute of Astronomy, National Tsing Hua University · Institute of Space Engineering, National Tsing Hua University

astro-ph.HE

Submitted: 2026-10-22

Updated: 2026-10-22

Comments: Accepted for publication in PASP; 17 pages, 2 figures, 9 tables

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

The gist: The gist: The study proposes a methodology to search for possible relations between gravitational waves (GWs) and Fast X-ray Transients (FXTs), utilizing the unmodeled search algorithm X-Pipeline to

Key concepts

Fast X-ray Transients (FXTs): These are brief, intense X-ray flashes lasting from seconds to hours without a persistent source. They are heterogeneous and have many possible origins, including outbursts from black holes or supernovae. The study uses them as potential triggers for searching for gravitational waves.
X-Pipeline: This is an unmodeled search algorithm used to find excess coherent signals in LIGO-Virgo-KAGRA data. It works by creating a time-frequency map, clustering the highest energy pixels, and then performing glitch rejection to identify potential candidate events.
Upper Limit Estimation: Since no definitive GW detection was found, the researchers calculated upper limits on the possible gravitational wave signals. This is done by injecting simulated waveforms into off-source data to determine the sensitivity of their search method against known signal types.

Summary

The gist: The study proposes a methodology to search for possible relations between gravitational waves (GWs) and Fast X-ray Transients (FXTs), utilizing the unmodeled search algorithm X-Pipeline to identify excess coherent signals in LIGO-Virgo-KAGRA (LVK) O4b observations

Introduction and Motivation

Fast X-Ray Transients (FXTs) are characterized by abrupt X-ray emissions spanning timescales of seconds to several hours and are spatially distinct from any persistent X-ray sources or previously identified stellar objects. FXTs represent a heterogeneous class of short-lived high-energy events, whose physical origins remain under active investigation. The observed diversity in peak luminosity, duration, spectral hardness, and host-galaxy environment suggests that multiple progenitor channels may contribute to the overall population. Major classes of proposed astrophysical origins include outbursts from X-ray binaries (XRBs), shock breakout of core-collapse supernovae (CCSNe), off-axis gamma-ray bursts (GRBs), tidal disruption events (TDEs) of systems with a white dwarf and an intermediate-mass black hole (IMBH), binary neutron star (BNS) mergers, proto-neutron stars in supernova models, collapsars in gravitationally unstable neutrino-cooled disks, and cocoon breakout caused by jet-envelope collisions.

Methodology for GW Search

The researchers intend to utilize X-Pipeline, an unmodeled search algorithm, to identify excess coherent signals in LIGO-Virgo-KAGRA (LVK) O4b observations. The general setup requires several user-defined preliminaries, including the set of detectors used (LIGO-Hanford (H1), LIGO-Livingston (L1), and Virgo (V1)), the range of the on-source window, data streams, sky positions, and numerical method parameters. For this search, an FXT event is analyzed if the burst occurred at all three detectors or at both H1 and L1. The on-source window is defined as [T0 − 600s, T0 + 60s], where T0 refers to the event time obtained from the Einstein Probe, based on the premise that FXTs and GRBs are driven by fundamentally similar astrophysical mechanisms.

Search Algorithm Details

The X-pipeline starts with the construction of a time-frequency map for each block, aiming to produce a coherent set of h× and h+ according to the maximum likelihood method. The intensity of the highest one percent of pixels would then be clustered and defined as a candidate event. Both on-source and off-source windows will be used to perform a search for candidate events. A glitch rejection process is performed on all on-source candidate events, using a threshold function defined by log10 I = a(log10 E)2 + c, where I and E represent the coherent energy and incoherent energy, respectively. The loudest event remaining after glitch rejection is called the loudest event. To claim that the loudest event is truly a GW detection, a p-value of the onsource loudest event compared to the background loud events must be calculated.

Upper Limit Estimation and Results

Since there is a high chance that the p-value of the loudest event does not imply detection of GW, an upper limit calculation is performed after estimating the probability. Simulated waveforms are injected into the off-source data, including Circular Sine–Gaussian (CSG), compact binary inspirals (BNS and NSBH mergers), and Accretion disk instability (ADI) signals. The sensitivity of the generic transient search is quantified by determining the percentage of injections that yield a significance higher than the loudest onsource event, which sets the upper limit. For notable sources like EP240413a, optical observations show it is located at a z= 0.41 galaxy, and the null detection of GW from this source is in agreement with an estimated p-value of 0.068.

Conclusion and Future Work

The X-pipeline results show that none of the FXTs have significant GW signals. A 90% confidence level exclusion distance is set for each source according to the upper limits of possible waveform models, which are listed in Table 2. The work demonstrates a steady sensitivity of NSBH but slightly weaker sensitivity of BNS in the search. Future studies could extend the on-source search window by considering other potential GW mechanisms, such as extending it by considering other potential GW mechanisms. Additionally, the time difference between the gamma-ray and X-ray triggers of a GRB might lead to a different on-source window. Furthermore, one could conduct a search using a single detector to expand the FXT sample or include KAGRA data to enhance sensitivity. The work will be carried out in the future as the observation of EP continues to capture FXTs with a high resolution and detection rate.

Table Summaries

Table 1 summarizes the final FXT list for analysis sorted in ascending order based on detection time. Table 2 presents the exclusion distances of FXT sources. Table 3 details the 90% upper limits on EGW,90% and hrss,90% for CSG and CBC waveforms for the search of EP240414a. Table 4 provides the upper limits on EGW,90% and hrss,90% for CSG and CBC waveforms for the search of EP250125a. Table 6 shows p-values estimated under the null hypothesis for all 19 sources.

References

Abbott, B. P., et al. 2017a, PhRvL, 119, 161101, doi: 10.1103/PhysRevLett.119.89

Abbott, B. P., et al. 2022, ApJ, 928, 186, doi: 10.3847/1538-4357/ac532b

Acernese, F., et al. 2022, Classical and Quantum Gravity, 39, 045006, doi: 10.1088/1361-6382/ac3c8e

Ai, S., et al. 2021, ApJL, 915, L11, doi: 10.3847/2041-8213/ac097d

Alp, D., et al. 2020, ApJ, 896, 39, doi: 10.3847/1538-4357/ab91ba

Aryan, A., et al. 2024a, GRB Coordinates Network, 36094, 1

Aryan, A., et al. 2024b, GRB Coordinates Network, 37575, 1

Aryan, A., et al. 2024c, GRB Coordinates Network, 37021, 1

Aryan, A., et al. 2025, GRB Coordinates Network, 39032, 1

Barria, P., et al. 2024, GRB Coordinates Network, 38556, 1

Bauer, F. E., et al. 2017, MNRAS, 467, 4841, doi: 10.1093/mnras/stx417

Belkin, S., & others. 2024, GRB Coordinates Network, 37583, 1

Brivio, R., et al. 2024, GRB Coordinates Network, 37579, 1

Bromberg, O., Tchekhovskoy, A., Gottlieb, O., Nakar, E., & Piran, T. 2018, MNRAS, 475, 2971

Bucciantini, N., Quataert, E., Metzger, B. D., et al. 2009, MNRAS, 396, 2038

Chen, W., et al. 2024a, GRB Coordinates Network, 36801, 1

Chen, W., et al. 2024b, GRB Coordinates Network, 36838, 1

Chen, W., et al.

Improvements for AI systems

Improved AI system can perform targeted search for gravitational wave counterparts to Fast X-ray Transients by utilizing an unmodeled search algorithm like X-pipeline to identify excess coherent signals in LIGO-Virgo-KAGRA (LVK) O4b observations.
Improved AI system can calculate upper limits and exclusion distances for each source based on simulated waveforms, using the methodology described as we select simulated waveforms that are physically motivated to produce both electromagnetic and GW emissions to inject into the off-source window.
Improved AI system can assess the statistical significance of potential GW detections by comparing the p-value of the onsource loudest event compared to the background loudest events, which gives the probability of such an event being ’not’ a GW detection, i.e., the probability under a null hypothesis.
Improved AI system can constrain gravitational wave emission efficiency for different waveform models by inverting Equation 2, as described in we can invert this relationship using Equation 2 (P. J. Sutton 2013) to establish a 90% upper limit on emission efficiency (EGW,90%).
Improved AI system can provide a statistical probability estimation for each FXT source by outputting the p-value, which shows values were compatible with the background fluctuations, indicating no GW events were found in our samples.