arXiv subjects

Qingtian Liu

Publications and source records attributed to Qingtian Liu.

4 recordsLinked to original sources

arXiv2026

Transit Searches for Habitable-Zone Exoplanets with Artificial Intelligence

Searching for Earth-sized planets in stellar habitable zones ultimately means finding a small number of very shallow, widely separated transits in noisy light curves. Around Sun-like stars, habitable-zone planets generally have long orbital periods, so a finite observing baseline may record only a few transits. At the same time, intrinsic stellar variability, instrumental systematics, and data gaps can obscure these weak events or produce transit-like brightness changes. This review examines recent progress in AI-assisted light-curve preprocessing, weak-signal searches, candidate vetting, and parameter inference, and considers the potential roles of pretraining and multimodal models. Taken together, the studies reviewed here show that AI has practical value in making large-scale light-curve analysis more efficient, recovering more signals at low signal-to-noise ratios in some injection tests, and reducing the number of targets requiring manual review. This review concludes that combining AI with physical modeling, statistical inference, and follow-up observations provides a practical route to improve both the discovery efficiency and the reliability of habitable-zone candidate studies.

astro-ph.EP↗

arXiv2026

DELOS: Contrastive Deep Learning for Low-SNR Blind Transit Searches in Kepler Photometry

We present DEtection in phase-folded Light curves with cOntrastive Scoring (DELOS), a deep-learning framework that uses contrastive scoring to perform blind searches for shallow transits in Kepler photometry. DELOS combines GPU-accelerated phase folding, optimized phase binning, and a custom one-dimensional convolutional encoder to assign a transit-likeness score to each folded light curve, thereby producing a score periodogram over trial periods without relying on pre-detected threshold-crossing events. Focusing on intermediate-to-long-period signals with orbital periods of 100-150 days, DELOS was trained on 20 million synthetic light curves generated with realistic transit models and Kepler-like noise properties, achieving a validation accuracy of 99.3% on the synthetic validation set. In controlled injection-recovery experiments, DELOS improves the combined precision-recall performance by 15.5% relative to Box-fitting Least Squares (BLS) and 11.25% relative to Transit Least Squares (TLS) in the low Signal-to-Noise Ratios (low-SNR) regime. It also accelerates the search by factors of approximately 3-5 and 74-80 compared with BLS and TLS, respectively. Applied to a selected Kepler validation sample, DELOS recovered all known shallow intermediate-to-long-period transit signals in the tested period range. These results demonstrate that DELOS provides an efficient and sensitive framework for low-SNR transit searches and represents a practical step toward future searches for longer-period terrestrial planets in Kepler, K2, TESS, PLATO, and Earth 2.0 data. Accordingly, this work is intended as a methodological development and validation study, with the detailed astrophysical validation of newly identified candidates deferred to future work.

astro-ph.EP↗

arXiv2026

TransitNet: A Compact Attention-Augmented Deep Learning Framework for Low-SNR Transit Blind Searches

Motivated by the observational incompleteness of intermediate-to-long-period Earth-size planets, we present TransitNet, a compact attention-augmented deep-learning framework for low-SNR transit blind searches. To enable realistic method development and objective threshold calibration under blind-search conditions, we develop a unified dataset construction, benchmarking, and threshold-selection framework. On recovery benchmarks constructed from unseen Kepler targets, TransitNet attains 95.2 percent accuracy in the challenging SNR range of 6 to 8 and outperforms both TLS and BLS, achieving ROC-AUC and PR-AP values of 0.974 and 0.982, respectively. In an injected Earth-size and sub-Earth-size transit recovery experiment, TransitNet achieves a recovery rate of 93.0 percent, substantially exceeding those of TLS (63.1 percent) and BLS (60.0 percent). In addition to detection, TransitNet provides attention-based estimates of transit windows and midpoints. On an independent evaluation set, 97.4 percent of injected transits are fully covered by the estimated transit window. Applied to real Kepler observations, the model successfully recovers all 34 selected confirmed Kepler planets, with a mean absolute transit midpoint error of 1.24 hours. The model combines a compact footprint of about 1.5 MB with high inference efficiency, yielding speed-ups of about 12 to 25 times relative to CPU-TLS and about 4 to 5 times relative to CPU-BLS. These results demonstrate that TransitNet provides an accurate, scalable, and computationally efficient framework for low-SNR transit blind searches in the tested regime and motivate its extension to longer-period Earth-size planet searches.

astro-ph.EP↗

arXiv2021

Physical Parameters of the Multi-Planet Systems HD 106315 and GJ 9827

HD 106315 and GJ 9827 are two bright, nearby stars that host multiple super-Earths and sub-Neptunes discovered by K2 that are well suited for atmospheric characterization. We refined the planets' ephemerides through Spitzer transits, enabling accurate transit prediction required for future atmospheric characterization through transmission spectroscopy. Through a multi-year high-cadence observing campaign with Keck/HIRES and Magellan/PFS, we improved the planets' mass measurements in anticipation of HST transmission spectroscopy. For GJ 9827, we modeled activity-induced radial velocity signals with a Gaussian process informed from the Calcium II H&K lines in order to more accurately model the effect of stellar noise on our data. We found planet masses of M$_b$=$4.87\pm 0.37$ M$_\oplus$, M$_c$=$1.92\pm 0.49$ M$_\oplus$, and M$_d$=$3.42\pm 0.62$ M$_\oplus$. For HD 106315, we found that such activity-radial velocity decorrelation was not effective due to the reduced presence of spots and speculate that this may extend to other hot stars as well (T$_{\rm {eff}}>6200$ K). We found planet masses of M$_b$=$10.5\pm 3.1$ M$_\oplus$ and M$_c$=$12.0\pm 3.8$ M$_\oplus$. We investigated all of the planets' compositions through comparing their masses and radii to a range of interior models. GJ 9827 b and GJ 9827 c are both consistent with an Earth-like rocky composition, GJ 9827 d and HD 106315 b both require additional volatiles and are consistent with moderate amounts of water or hydrogen/helium, and HD 106315 c is consistent with 10% hydrogen/helium surrounding an Earth-like rock and iron core.

astro-ph.EP↗