REVIEW
Search for high-energy neutrinos from Einstein Probe detected Fast X-ray Transients using 14 years of IceCube muon track data
High Energy Astrophysical Phenomena · Submitted 22 Oct 2026 · 49 references
Pith's verdict: Clean, honest null result for neutrinos from Einstein Probe FXTs, but the temporal mismatch between the IceCube data window (2008–2022) and the FXT triggers (2024–2026) means the search only constrains steady emission, not the most plausible transient models.
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
A search for high-energy neutrinos from Einstein Probe fast X-ray transients, using 14 years of IceCube muon-track data, finds no statistically significant excess from any of the 27 sources.
The reading
What carries the argument
The unbinned maximum likelihood method (Braun et al. 2008) as implemented in the SkyLLH software is the core statistical tool; it constructs a likelihood ratio test statistic by modelling the spatial and energy distribution of background (from data scrambling) and signal (a power-law spectrum with free spectral index). The signal PDF combines a spatial term (the point-spread function of IceCube) and an energy term derived from detector effective areas. The test statistic is maximized over the number of signal events $n_s$ and spectral index $\gamma$ for each source, and background trials are used to compute local p-values. The analysis uses the public IceTracks-DR2 dataset (1.64 million neutrino events) and a catalog of 27 FXTs assembled from O'Connor et al. (2025) plus EP260119a.
Extended reading notes
Core claim
The central claim is that no high-energy neutrino point source coincident with any of the 27 spectroscopically confirmed fast X-ray transients from Einstein Probe is detected in a time-integrated search using 14 years of IceCube muon-track data. The maximum test statistic (TS = 4.77 for EP240801a) corresponds to a local p-value of 0.04 and is not statistically significant. The 90% confidence level upper limits on the differential muon-neutrino flux at 100 TeV (assuming an $E^{-2}$ spectrum) range from $2.1\times10^{-10}$ to $6.1\times10^{-9}$ GeV cm$^{-2}$ s$^{-1}$. These limits are roughly an order of magnitude higher than the predicted neutrino rate from the choked jet model for a source at $z=1$ with $E_{\rm iso}\sim10^{53}$ erg, so the model is not ruled out.
Load-bearing premise
The analysis assumes that a time-integrated search over 14 years, with no temporal overlap between the IceCube data (ending 2022) and the FXT trigger times (2024–2026), is sensitive enough to detect neutrino emission from FXTs whose predicted signal is transient and coincident with the X-ray event.
Editorial extensions
If this is right
- If the choked-jet model for FXTs is correct and produces neutrinos in coincidence with the X-ray burst, the current time-integrated limits are not constraining; a time-resolved search with IceCube data overlapping the FXT trigger periods is required to test the model.
- The most stringent upper limit (EP241021a, $E^2\Phi(E)=2.1\times10^{-10}$ GeV cm$^{-2}$ s$^{-1}$ at 100 TeV) can be used to normalize population models for FXT neutrino emission under the assumption of a steady flux.
- For the 15 FXTs that returned a non-zero test statistic, the best-fit spectral indices are soft ($\gamma>2.9$), suggesting that if a signal were present, it would be at lower energies and therefore harder to detect above atmospheric backgrounds.
- The absence of a significant excess across 27 sources places an upper bound on the fraction of FXTs that could be steady neutrino sources at the level of $\sim$10% of the choked-jet prediction per source, assuming the model parameters used in Mészáros & Waxman (2001).
Reading between the lines
- Because the FXTs were detected after the IceCube data collection window, the analysis is effectively a stacked search for steady emission from an ensemble of sources that are individually variable. Any transient neutrino flux coincident with the X-ray event is diluted over 14 years; the limits therefore mainly constrain persistent or long-lived emission, not the choked-jet scenario the authors aim
- The fact that the largest TS occurs for EP240801a, which has a redshift $z=1.673$ and no detected GRB counterpart, could be a statistical fluctuation; but if future time-domain IceCube observations of this source become available, it merits a targeted temporal search.
- A joint analysis combining IceCube's cascade channel (sensitive to all flavours) with muon tracks, and using the full 10 years of overlapping data after 2022 (once released), would improve sensitivity to transient neutrino signals from the many new FXTs expected from Einstein Probe's ongoing survey.
Original abstract
We perform a time-integrated search for high energy neutrinos from 27 fast X-ray transients (FXTs) detected by Einstein Probe, using 14 years of publicly available IceCube muon-track data. We employ the unbinned maximum likelihood method which is used for IceCube point source searches. We do not find any statistically significant excess from any of the FXTs. The largest TS value is obtained for EP240801a (TS=4.77), corresponding to a local $p$-value of 0.04, which is not significant. We therefore calculate 90\% confidence level upper limit on the differential muon neutrino flux. The most stringent upper limit is obtained for EP241021a, with $E^2 \Phi(E)=2.1\times 10^{-10}$ (GeV cm$^{-2}$ s$^{-1}$), for a $E^{-2}$ spectrum at a pivot energy of 100 TeV.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
We perform a time-integrated search for high energy neutrinos from 27 fast X-ray transients (FXTs) detected by Einstein Probe, using 14 years of publicly available IceCube muon-track data. We employ the unbinned maximum likelihood method which is used for IceCube point source searches. We do not find any statistically significant excess from any of the FXTs. The largest TS value is obtained for EP240801a (TS=4.77), corresponding to a local $p$-value of 0.04, which is not significant. We therefore calculate 90\% confidence level upper limit on the differential muon neutrino flux. The most stringent upper limit is obtained for EP241021a, with $E^2 \Phi(E)=2.1\times 10^{-10}$ (GeV cm$^{-2}$ s$^{-1}$), for a $E^{-2}$ spectrum at a pivot energy of 100 TeV.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | We perform a time-integrated search for high energy neutrinos from 27 fast X-ray transients (FXTs) detected by Einstein Probe, using 14 years of publicly available IceCube muon-track data. | contribution 0.92 | 0.96 | 0.05 | asserted 0.95 |
| s1 | We employ the unbinned maximum likelihood method which is used for IceCube point source searches. | method 1.00 | 0.75 | 0.01 | asserted 1.00 |
| s2 | We do not find any statistically significant excess from any of the FXTs. | result 1.00 | 0.97 | 0.89 | asserted 1.00 |
| s3 | The largest TS value is obtained for EP240801a (TS=4.77), corresponding to a local $p$-value of 0.04, which is not significant. | result 0.95 | 0.97 | 0.25 | asserted 1.00 |
| s4 | We therefore calculate 90\% confidence level upper limit on the differential muon neutrino flux. | contribution 0.43 | 0.96 | 0.07 | asserted 0.98 |
| s5 | The most stringent upper limit is obtained for EP241021a, with $E^2 \Phi(E)=2.1\times 10^{-10}$ (GeV cm$^{-2}$ s$^{-1}$), for a $E^{-2}$ spectrum at a pivot energy of 100 TeV. | result 1.00 | 0.96 | 0.17 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v1
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
Figures
Figures reproduced from arXiv:2610.30722 by the authors.
References
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.