pith. sign in

arxiv: 2605.23739 · v1 · pith:RCL7LDUOnew · submitted 2026-05-22 · 🌌 astro-ph.IM

A Wavelet-Integrated Search Pipeline for Narrowband Technosignatures in FAST Observations of 33 Exoplanet Systems

Zi-Qi Li , Jian-Kang Li , Zhe-Wei Luo , Chen-Xu Guan , Yu-Tong Fan , Zhen-Zhao Tao , Xiao-Hang Luan , Bo-Lun Huang
show 9 more authors
Yu Hu Peng-Yu Li Pu-Fan Liu Kang Jiao Tong-Jie Zhang Hai-Yan Zhang Peng Jiang Rui Li Liang Gao
This is my paper

Pith reviewed 2026-05-25 02:48 UTC · model grok-4.3

classification 🌌 astro-ph.IM
keywords narrowband technosignatureswavelet analysisFAST telescopeSETIexoplanet observationsradio dynamic spectrasignal detectionMSWNet
0
0 comments X

The pith

A wavelet-integrated pipeline detects narrowband technosignatures in FAST data by extracting multi-scale features and regressing endpoints.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents a new search method for narrowband signals that might indicate technology in observations of 33 exoplanet systems taken with the FAST radio telescope. Instead of using simple intensity thresholds on drifting signals, it employs a Multi-Scale Wavelet Net to break the data into different resolution levels and then uses regression to pinpoint the start and end of potential signals. Additional steps filter based on signal shape, check signal strength in raw data, and use multiple beams to reject interference. When run on the actual observations, this pipeline successfully finds signals noted in earlier studies and outputs a small number of candidates that can be checked further. The design keeps the steps understandable and allows easy adjustment of detection limits.

Core claim

The pipeline reframes narrowband detection as wavelet-guided feature extraction followed by endpoint regression, morphology-aware filtering, raw-data S/N validation, and multi-beam anticoincidence veto. Applied to real FAST data on 33 exoplanet systems, it recovers representative events from prior analyses and produces a compact set of veto-ready candidates.

What carries the argument

Multi-Scale Wavelet Net (MSWNet) for producing an interpretable multi-resolution representation, followed by a lightweight parameter estimator for endpoint localization.

If this is right

  • The pipeline recovers representative events from prior analyses on the FAST data.
  • It produces a compact set of veto-ready candidates for downstream inspection.
  • The workflow preserves interpretability, low regression complexity, and auditable threshold control.
  • It is readily transferable to other radio surveys and large-scale technosignature searches.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The method may scale better to larger volumes of data from upcoming surveys by reducing the number of candidates needing manual review.
  • It could be adapted to search for other types of signals beyond narrowband ones if the wavelet features capture the morphology.
  • Testing on simulated datasets with varying noise levels would confirm if sensitivity is maintained across different conditions.

Load-bearing premise

That reframing narrowband detection as wavelet-guided feature extraction followed by endpoint regression maintains or improves sensitivity without introducing new classes of missed signals or false positives.

What would settle it

Running both the new pipeline and a traditional hard-threshold drift search on the same FAST dataset and comparing the number of recovered known signals and the rate of new false candidates.

Figures

Figures reproduced from arXiv: 2605.23739 by Bo-Lun Huang, Chen-Xu Guan, Hai-Yan Zhang, Jian-Kang Li, Kang Jiao, Liang Gao, Peng Jiang, Peng-Yu Li, Pu-Fan Liu, Rui Li, Tong-Jie Zhang, Xiao-Hang Luan, Yu Hu, Yu-Tong Fan, Zhen-Zhao Tao, Zhe-Wei Luo, Zi-Qi Li.

Figure 1
Figure 1. Figure 1: Schematic architecture of MSWNet. The overall encoder–decoder structure is summarized in the left panel, while the right panel expands a single stage showing the internal encoder/decoder layout, where all resolution changes are performed within the blocks. The encoder replaces pooling with DWT2D, caching detail coefficients (LH, HL, HH) at each level, which are then reused by the decoder through IDWT2D tog… view at source ↗
Figure 2
Figure 2. Figure 2: Inference pipeline. MSWNet transforms each raw patch into a processed patch (cleaned map), and a lightweight parameter estimator outputs a fixed set of predictions. Predictions are gated by a patch-scale global SNR statistic, filtered by confidence and IoU (NMS), stitched across overlapping patches, and then validated by S/N measured on the raw spectrogram. Events surviving the veto stage constitute the fi… view at source ↗
Figure 3
Figure 3. Figure 3: Sensitivity of detection yield to post-processing thresholds. Each panel shows a heat map of the retained fraction of detections versus the confidence cutoff and the NMS IoU threshold, under different Global-SNR activation settings. The frequency histograms show candidates distributed across the full 1.05–1.45 GHz band, with prominent clus￾tering at known interference sub-bands rather than any preference f… view at source ↗
Figure 4
Figure 4. Figure 4: Distributions of detected-signal parameters at different pipeline stages. Histograms show detections, events, and final candidates as functions of observing frequency, drift rate, S/N measured on raw data, Global SNR on the cleaned map, and model confidence. decisive discriminator, a single-feed response is a com￾mon warning sign of instrumental or terrestrial contam￾ination(J. Tinbergen 1996). The measure… view at source ↗
Figure 5
Figure 5. Figure 5: Dynamic spectrum (frequency-time waterfall) from the 19-beam L-band receiver during the K2-155 observation, centered on the candidate frequency. The narrow drifting signal (1148.4167 MHz, drift −0.038 Hz/s) appears in Beam 1 (on target) and is not detected in any of the other 18 beams. narrowband features dominated by Y Y near the same frequency strongly favors a common interference origin over a signal lo… view at source ↗
Figure 6
Figure 6. Figure 6: Comparison of the 1148.4167 MHz region in XX vs. YY polarization for four targets observed on 2021-09-10 (arranged by time). short pulsed bursts or pulse pairs (R. J. Kerczewski et al. 2015) and therefore do not naturally repro￾duce a smooth, continuous drifting track, making them a weak match to Condition1 despite being in-band. Military emitters such as Link-16 and many L-band radars are likewise typical… view at source ↗
Figure 7
Figure 7. Figure 7: Ensemble diagnostics for the matched-control hits. (a) Frequency–S/N distribution of the 201 matched hits, colored by group. (b) Rayleigh-style spacing scan, with the strongest response at ∆ν ∗ ≃ 12.344 kHz. (c) Wrapped residual/phase structure computed at ∆ν ∗ with f0 = min(fi), showing that the locked band extends across the survey frequency range. ily, rather than making it a typical member of that inte… view at source ↗
Figure 8
Figure 8. Figure 8: Inference-time wavelet-path ablation. From left to right: denoised output, LL-only output, inverse-synthe￾sized feature, and its LL-only counterpart. To assess the role of the detail pathway, we performed an inference-time ablation: all encoder-to-decoder high￾frequency transfers were set to zero, forcing the recon￾struction to rely on the low-frequency pathway alone. We find that in low-S/N regimes, parti… view at source ↗
Figure 9
Figure 9. Figure 9: Robustness gallery (left to right). Yellow boxes denote detections morphologically classified as linear, while red boxes indicate signals with curvature. (a) Target drifting narrowband track + broadband impulsive transient, with the corresponding MSWNet cleaned map and predicted box. (b) Target + curved interfering track, with the cleaned map and predicted box. (c) Target + quasi-sinusoidally modulated int… view at source ↗
Figure 10
Figure 10. Figure 10: Representative false positives for the incoherent de-Doppler drift search (left; Taylor-tree implementation in TurboSETI) and MSWNet pipeline (right). a limitation of MSWNet’s upstream feature extraction; in future work, we will explore replacing (or augment￾ing) the estimator with physics-driven track fitting (e.g., Hough-transform-based line/curve detection) to achieve higher localization precision whil… view at source ↗
read the original abstract

Building on prior FAST targeted and blind SETI campaigns toward 33 exoplanet systems, we introduce a wavelet-integrated search pipeline for narrowband technosignature candidates in radio dynamic spectra. At its core, the pipeline uses a Multi-Scale Wavelet Net (MSWNet) to produce an interpretable multi-resolution representation, followed by a lightweight parameter estimator for endpoint localization. Rather than relying solely on hard-threshold drift searches, the pipeline reframes narrowband detection as wavelet-guided feature extraction followed by endpoint regression, morphology-aware filtering, raw-data S/N validation, and multi-beam anticoincidence veto. Applied to real FAST data, the pipeline recovers representative events from prior analyses and produces a compact set of veto-ready candidates for downstream inspection. The resulting workflow preserves interpretability, low regression complexity, and auditable threshold control, making it readily transferable to other radio surveys and large-scale technosignature searches.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript presents a wavelet-integrated search pipeline for narrowband technosignatures in FAST radio dynamic spectra toward 33 exoplanet systems. At its core is the Multi-Scale Wavelet Net (MSWNet) for multi-resolution feature extraction, followed by lightweight endpoint regression, morphology-aware filtering, raw-data S/N validation, and multi-beam anticoincidence veto. Applied to real FAST data, the pipeline is stated to recover representative events from prior analyses while producing a compact set of veto-ready candidates, with emphasis on preserved interpretability, low regression complexity, and auditable thresholds.

Significance. If the central performance claims hold, the work would supply a transferable, interpretable alternative to hard-threshold drift searches for narrowband SETI, building directly on existing FAST campaigns. The emphasis on auditable thresholds and morphology-aware steps is a constructive addition for downstream human inspection.

major comments (2)
  1. [Abstract and §3] Abstract and §3 (pipeline core): The claim that reframing detection as MSWNet-guided extraction plus endpoint regression 'preserves' sensitivity relative to hard-threshold drift searches is load-bearing for the central result, yet no injection-recovery statistics, completeness curves, or direct false-alarm rate comparisons are supplied; recovery of 'representative events' alone does not constrain systematic misses or localization bias at low S/N.
  2. [§4] §4 (real-data application): The statement that the pipeline 'recovers representative events from prior analyses' requires explicit quantitative metrics (e.g., recovery fraction, S/N distribution of recovered vs. missed events, or overlap table with prior candidate lists) to be evaluable; without these the downstream claim of a 'compact set of veto-ready candidates' cannot be assessed for completeness.
minor comments (2)
  1. [§3.1] Notation for MSWNet output dimensions and the regression loss function should be defined explicitly in §3.1 before use in later equations.
  2. [Figures] Figure captions for dynamic spectra should state the exact frequency resolution, time resolution, and beam configuration used in the displayed examples.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive review. We address each major comment below, clarifying the manuscript's claims and indicating revisions where appropriate.

read point-by-point responses

  1. Referee: [Abstract and §3] Abstract and §3 (pipeline core): The claim that reframing detection as MSWNet-guided extraction plus endpoint regression 'preserves' sensitivity relative to hard-threshold drift searches is load-bearing for the central result, yet no injection-recovery statistics, completeness curves, or direct false-alarm rate comparisons are supplied; recovery of 'representative events' alone does not constrain systematic misses or localization bias at low S/N.

    Authors: The manuscript does not claim that sensitivity is preserved relative to hard-threshold drift searches. The abstract and §3 state that the workflow preserves interpretability, low regression complexity, and auditable threshold control. Recovery of representative events is presented to demonstrate applicability to real FAST data rather than to quantify completeness. We agree that injection-recovery statistics would strengthen the work and have added a limitations paragraph in the discussion section noting the absence of such tests and the scope for future injection studies. revision: partial

  2. Referee: [§4] §4 (real-data application): The statement that the pipeline 'recovers representative events from prior analyses' requires explicit quantitative metrics (e.g., recovery fraction, S/N distribution of recovered vs. missed events, or overlap table with prior candidate lists) to be evaluable; without these the downstream claim of a 'compact set of veto-ready candidates' cannot be assessed for completeness.

    Authors: We have revised §4 to include an overlap table with the prior candidate lists referenced in the text, along with the recovery fraction for the representative events. The S/N values of recovered events are now explicitly listed. A full accounting of missed events at low S/N is not feasible from the information provided in the cited prior works, which do not publish exhaustive candidate lists; this limitation is now stated in the revised text. revision: yes

Circularity Check

0 steps flagged

No circularity: new pipeline is independent addition with no self-referential reductions.

full rationale

The manuscript introduces MSWNet multi-resolution feature extraction followed by endpoint regression and morphology-aware filtering as a distinct workflow reframing narrowband detection. This is applied to real FAST data to recover representative prior events and generate veto-ready candidates. No equations, parameters, or performance claims reduce by construction to fitted inputs, self-citations, or renamed prior results. Prior FAST campaign citations provide context but are not load-bearing for the new pipeline's claimed properties of interpretability and auditable thresholds. The derivation chain remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

Review is limited to the abstract; the central claim rests on the unexamined effectiveness of the MSWNet for the stated task and on the assumption that the listed processing steps do not degrade detection performance.

invented entities (1)
  • Multi-Scale Wavelet Net (MSWNet) no independent evidence
    purpose: Produce interpretable multi-resolution representation for narrowband feature extraction and endpoint localization
    Introduced in the abstract as the core component of the new pipeline.

pith-pipeline@v0.9.0 · 5757 in / 1209 out tokens · 46627 ms · 2026-05-25T02:48:09.547237+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

52 extracted references · 52 canonical work pages

  1. [1]

    2017, Astronomy and Computing, 18, 35, doi: 10.1016/j.ascom.2017.01.002

    Akeret, J., Chang, C., Lucchi, A., & Refregier, A. 2017, Astronomy and Computing, 18, 35, doi: 10.1016/j.ascom.2017.01.002

    work page doi:10.1016/j.ascom.2017.01.002 2017

  2. [2]

    2002, Communications of the ACM, 45, 56, doi: 10.1145/581571.581573

    Werthimer, D. 2002, Communications of the ACM, 45, 56, doi: 10.1145/581571.581573

    work page doi:10.1145/581571.581573 2002

  3. [3]

    Bodla, N., Singh, B., Chellappa, R., & Davis, L. S. 2017, in 2017 IEEE International Conference on Computer Vision (ICCV), 5562–5570, doi: 10.1109/ICCV.2017.593

    work page doi:10.1109/iccv.2017.593 2017

  4. [4]

    Brzycki, B., Siemion, A. P. V., Pater, I. d., et al. 2022, AJ, 163, 222, doi: 10.3847/1538-3881/ac5e3d

    work page doi:10.3847/1538-3881/ac5e3d 2022

  5. [5]

    1959, Nature, 184, 844, doi: 10.1038/184844a0

    Cocconi, G., & Morrison, P. 1959, Nature, 184, 844, doi: 10.1038/184844a0

    work page doi:10.1038/184844a0 1959

  6. [6]

    2024, A&A, 690, A211, doi: 10.1051/0004-6361/202449548

    Cornu, D., Salom´ e, P., Semelin, B., et al. 2024, A&A, 690, A211, doi: 10.1051/0004-6361/202449548

    work page doi:10.1051/0004-6361/202449548 2024

  7. [7]

    Drake, F. D. 1961, Physics Today, 14, 40, doi: 10.1063/1.3057500 ECC (CEPT). 2011, Impact of Unwanted Emissions of Iridium Satellites on Radio Astronomy Operations in the Band 1610.6–1613.8 MHz, ECC Report 171, Electronic Communications Committee (ECC), CEPT, Tallinn, Estonia. https://docdb.cept.org/document/279

    work page doi:10.1063/1.3057500 1961

  8. [8]

    E., Siemion, A., Foster, G., et al

    Enriquez, J. E., Siemion, A., Foster, G., et al. 2017, ApJ, 849, 104, doi: 10.3847/1538-4357/aa8d1b

    work page doi:10.3847/1538-4357/aa8d1b 2017

  9. [9]

    2023, A&A, 677, A101, doi: 10.1051/0004-6361/202345976

    Grishin, K., Mei, S., & Ili´ c, S. 2023, A&A, 677, A101, doi: 10.1051/0004-6361/202345976

    work page doi:10.1051/0004-6361/202345976 2023

  10. [10]

    2022, IEEE Access, 10, 58869, doi: 10.1109/ACCESS.2022.3179517

    Guo, T., Zhang, T., Lim, E., et al. 2022, IEEE Access, 10, 58869, doi: 10.1109/ACCESS.2022.3179517

    work page doi:10.1109/access.2022.3179517 2022

  11. [11]

    Harp, G. R. 2005, Radio Science, 40, RS5S18, doi: 10.1029/2004RS003133

    work page doi:10.1029/2004rs003133 2005

  12. [12]

    R., Richards, J., Tarter, J

    Harp, G. R., Richards, J., Tarter, J. C., et al. 2016, AJ, 152, 181, doi: 10.3847/0004-6256/152/6/181

    work page doi:10.3847/0004-6256/152/6/181 2016

  13. [13]

    H., et al

    Hirano, T., Dai, F., Livingston, J. H., et al. 2018, AJ, 155, 124, doi: 10.3847/1538-3881/aaaa6e 18

    work page doi:10.3847/1538-3881/aaaa6e 2018

  14. [14]

    2026, AJ, 171, 51, doi: 10.3847/1538-3881/ae2470

    Huang, B.-L., Tao, Z.-Z., Zhang, T.-J., & Gajjar, V. 2026, AJ, 171, 51, doi: 10.3847/1538-3881/ae2470

    work page doi:10.3847/1538-3881/ae2470 2026

  15. [15]

    2019, Science China

    Jiang, P., Yue, Y., Gan, H., et al. 2019, Science China

    work page 2019

  16. [16]

    Physics, Mechanics, and Astronomy, 62, 959502, doi: 10.1007/s11433-018-9376-1

    work page doi:10.1007/s11433-018-9376-1

  17. [17]

    A., & Wilson, J

    Shalkhauser, K. A., & Wilson, J. D. 2015, in 2015 IEEE Aerospace Conference, 1–12, doi: 10.1109/AERO.2015.7119060

    work page doi:10.1109/aero.2015.7119060 2015

  18. [18]

    2018, IEEE Microw

    Li, D., Wang, P., Qian, L., et al. 2018, IEEE Microwave Magazine, 19, 112, doi: 10.1109/MMM.2018.2802178

    work page doi:10.1109/mmm.2018.2802178 2018

  19. [19]

    2020, Research in Astronomy and Astrophysics, 20, 078, doi: 10.1088/1674-4527/20/5/78

    Li, D., Gajjar, V., Wang, P., et al. 2020, Research in Astronomy and Astrophysics, 20, 078, doi: 10.1088/1674-4527/20/5/78

    work page doi:10.1088/1674-4527/20/5/78 2020

  20. [20]

    2024, AJ, 167, 8, doi: 10.3847/1538-3881/ad0be8

    Li, J.-K., Chen, Y., Huang, B.-L., et al. 2024, AJ, 167, 8, doi: 10.3847/1538-3881/ad0be8

    work page doi:10.3847/1538-3881/ad0be8 2024

  21. [21]

    2022, ApJ, 938, 1, doi: 10.3847/1538-4357/ac90bd

    Li, J.-K., Zhao, H.-C., Tao, Z.-Z., Zhang, T.-J., & Xiao-Hui, S. 2022, ApJ, 938, 1, doi: 10.3847/1538-4357/ac90bd

    work page doi:10.3847/1538-4357/ac90bd 2022

  22. [22]

    2021, IEEE Transactions on Image Processing, 30, 7074, doi: 10.1109/TIP.2021.3101395

    Li, Q., Shen, L., Guo, S., & Lai, Z. 2021, IEEE Transactions on Image Processing, 30, 7074, doi: 10.1109/TIP.2021.3101395

    work page doi:10.1109/tip.2021.3101395 2021

  23. [23]

    2025, Frontiers in Astronomy and Space Sciences, Volume 12 - 2025, doi: 10.3389/fspas.2025.1656917

    Liu, P., Li, H., Li, Z., et al. 2025, Frontiers in Astronomy and Space Sciences, Volume 12 - 2025, doi: 10.3389/fspas.2025.1656917

    work page doi:10.3389/fspas.2025.1656917 2025

  24. [24]

    2023, AJ, 165, 132, doi: 10.3847/1538-3881/acb706

    Luan, X.-H., Tao, Z.-Z., Zhao, H.-C., et al. 2023, AJ, 165, 132, doi: 10.3847/1538-3881/acb706

    work page doi:10.3847/1538-3881/acb706 2023

  25. [25]

    MacMahon, D. H. E., Price, D. C., Lebofsky, M., et al. 2018, PASP, 130, 044502, doi: 10.1088/1538-3873/aa80d2

    work page doi:10.1088/1538-3873/aa80d2 2018

  26. [26]

    Nieuwpoort, R. V. 2022, MNRAS, 516, 5367, doi: 10.1093/mnras/stac2503

    work page doi:10.1093/mnras/stac2503 2022

  27. [27]

    2011, Int

    Nan, R., Li, D., Jin, C., et al. 2011, International Journal of Modern Physics D, 20, 989, doi: 10.1142/S0218271811019335

    work page doi:10.1142/s0218271811019335 2011

  28. [28]

    Neubeck and L

    Neubeck, A., & Van Gool, L. 2006, in 18th International Conference on Pattern Recognition (ICPR’06), Vol. 3, 850–855, doi: 10.1109/ICPR.2006.479

    work page doi:10.1109/icpr.2006.479 2006

  29. [29]

    J., Reeves J

    Offringa, A. R., de Bruyn, A. G., Biehl, M., et al. 2010, MNRAS, 405, 155, doi: 10.1111/j.1365-2966.2010.16471.x

    work page doi:10.1111/j.1365-2966.2010.16471.x 2010

  30. [30]

    C., Enriquez, J

    Price, D. C., Enriquez, J. E., Brzycki, B., et al. 2020, AJ, 159, 86, doi: 10.3847/1538-3881/ab65f1

    work page doi:10.3847/1538-3881/ab65f1 2020

  31. [31]

    2020, The Innovation, 1, 100053, doi: 10.1016/j.xinn.2020.100053

    Qian, L., Yao, R., Sun, J., et al. 2020, The Innovation, 1, 100053, doi: 10.1016/j.xinn.2020.100053

    work page doi:10.1016/j.xinn.2020.100053 2020

  32. [32]

    2016, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 779–788, doi: 10.1109/CVPR.2016.91

    Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. 2016, in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 779–788, doi: 10.1109/CVPR.2016.91

    work page doi:10.1109/cvpr.2016.91 2016

  33. [33]

    Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,

    Ren, S., He, K., Girshick, R., & Sun, J. 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence, 39, 1137, doi: 10.1109/TPAMI.2016.2577031

    work page doi:10.1109/tpami.2016.2577031 2017

  34. [34]

    2014, Proceedings of the National Academy of Science, 111, 12634, doi: 10.1073/pnas.1304213111

    Seager, S. 2014, Proceedings of the National Academy of Science, 111, 12634, doi: 10.1073/pnas.1304213111

    work page doi:10.1073/pnas.1304213111 2014

  35. [35]

    Z., Wright, J

    Sheikh, S. Z., Wright, J. T., Siemion, A., & Enriquez, J. E. 2019, ApJ, 884, 14, doi: 10.3847/1538-4357/ab3fa8

    work page doi:10.3847/1538-4357/ab3fa8 2019

  36. [36]

    Z., Smith, S., Price, D

    Sheikh, S. Z., Smith, S., Price, D. C., et al. 2021, Nature Astronomy, 5, 1153, doi: 10.1038/s41550-021-01508-8

    work page doi:10.1038/s41550-021-01508-8 2021

  37. [37]

    Siemion, A. P. V., Demorest, P., Korpela, E., et al. 2013, ApJ, 767, 94, doi: 10.1088/0004-637X/767/1/94

    work page doi:10.1088/0004-637x/767/1/94 2013

  38. [38]

    C., Sheikh, S

    Smith, S., Price, D. C., Sheikh, S. Z., et al. 2021, Nature Astronomy, 5, 1148, doi: 10.1038/s41550-021-01479-w

    work page doi:10.1038/s41550-021-01479-w 2021

  39. [39]

    2023, AJ, 166, 190, doi: 10.3847/1538-3881/acfc1e

    Tao, Z.-Z., Huang, B.-L., Luan, X.-H., et al. 2023, AJ, 166, 190, doi: 10.3847/1538-3881/acfc1e

    work page doi:10.3847/1538-3881/acfc1e 2023

  40. [40]

    2022, AJ, 164, 160, doi: 10.3847/1538-3881/ac8bd5

    Tao, Z.-Z., Zhao, H.-C., Zhang, T.-J., et al. 2022, AJ, 164, 160, doi: 10.3847/1538-3881/ac8bd5

    work page doi:10.3847/1538-3881/ac8bd5 2022

  41. [41]

    2001, ARA&A, 39, 511, doi: 10.1146/annurev.astro.39.1.511

    Tarter, J. 2001, ARA&A, 39, 511, doi: 10.1146/annurev.astro.39.1.511

    work page doi:10.1146/annurev.astro.39.1.511 2001

  42. [42]

    1996, in Astronomical Polarimetry (Cambridge University Press), 1–7, doi: 10.1017/CBO9780511525100.002

    Tinbergen, J. 1996, in Astronomical Polarimetry (Cambridge University Press), 1–7, doi: 10.1017/CBO9780511525100.002

    work page doi:10.1017/cbo9780511525100.002 1996

  43. [43]

    2021, AJ, 161, 286, doi: 10.3847/1538-3881/abf649 van Heerden, E., Karastergiou, A., & Roberts, S

    Traas, R., Croft, S., Gajjar, V., et al. 2021, AJ, 161, 286, doi: 10.3847/1538-3881/abf649 van Heerden, E., Karastergiou, A., & Roberts, S. J. 2017, MNRAS, 467, 1661, doi: 10.1093/mnras/stw3068

    work page doi:10.3847/1538-3881/abf649 2021

  44. [44]

    2021, Research in Astronomy and Astrophysics, 21, 018, doi: 10.1088/1674-4527/21/1/18

    Wang, Y., Zhang, H.-Y., Hu, H., et al. 2021, Research in Astronomy and Astrophysics, 21, 018, doi: 10.1088/1674-4527/21/1/18

    work page doi:10.1088/1674-4527/21/1/18 2021

  45. [45]

    S., et al

    Werthimer, D., Anderson, D., Bowyer, C. S., et al. 2001, in Society of Photo-Optical Instrumentation Engineers (SPIE) Conference Series, Vol. 4273, The Search for Extraterrestrial Intelligence (SETI) in the Optical Spectrum III, ed. S. A. Kingsley & R. Bhathal, 104–109, doi: 10.1117/12.435384

    work page doi:10.1117/12.435384 2001

  46. [46]

    2018, in International Conference on Learning Representations

    Williams, T., & Li, R. 2018, in International Conference on Learning Representations. https://openreview.net/forum?id=rkhlb8lCZ

    work page 2018

  47. [47]

    2021, Research in Astronomy and Astrophysics, 21, 119, doi: 10.1088/1674-4527/21/5/119

    Yan, R.-Q., Dai, C., Liu, W., et al. 2021, Research in Astronomy and Astrophysics, 21, 119, doi: 10.1088/1674-4527/21/5/119

    work page doi:10.1088/1674-4527/21/5/119 2021

  48. [48]

    B., & G¨ und¨ uz, D

    Yang, Q., Mashhadi, M. B., & G¨ und¨ uz, D. 2019, in 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP), 1–6, doi: 10.1109/MLSP.2019.8918798

    work page doi:10.1109/mlsp.2019.8918798 2019

  49. [49]

    2025, A&A, 701, A81, doi: 10.1051/0004-6361/202554423

    You, Z.-Y., Luo, J.-G., Pan, Y.-R., et al. 2025, A&A, 701, A81, doi: 10.1051/0004-6361/202554423

    work page doi:10.1051/0004-6361/202554423 2025

  50. [50]

    2024, in Journal of Physics Conference Series, Vol

    Zhang, T.-J., Huang, B.-L., Li, J.-K., et al. 2024, in Journal of Physics Conference Series, Vol. 2877, Journal of Physics Conference Series (IOP), 012053, doi: 10.1088/1742-6596/2877/1/012053 19

    work page doi:10.1088/1742-6596/2877/1/012053 2024

  51. [51]

    2024, Astronomy and Computing, 47, 100822, doi: 10.1016/j.ascom.2024.100822

    Zhang, X., Cognard, I., & Dobigeon, N. 2024, Astronomy and Computing, 47, 100822, doi: 10.1016/j.ascom.2024.100822

    work page doi:10.1016/j.ascom.2024.100822 2024

  52. [52]

    2020, Journal of Deep Space Exploration, 7, 158, doi: 10.15982/j.issn.2095-7777.2020.20190618005

    Zhisong, Z., Haiyan, Z., Yan, Z., et al. 2020, Journal of Deep Space Exploration, 7, 158, doi: 10.15982/j.issn.2095-7777.2020.20190618005

    work page doi:10.15982/j.issn.2095-7777.2020.20190618005 2020