---
格式版本: 2
标题: "Astro-Hunters: Machine Learning for Exoplanet Transit Detection in TESS Photometry"
原文链接: "https://arxiv.org/abs/2608.18172"
发布日期: "2026-08-17"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Mon, 17 Aug 2026 18:41:30 UTC (205 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-08-20T15:53:09+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-08-20T15:48:16+08:00"
入库时间: "2026-08-20T07:53:09.243Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=model%20fit&searchtype=all"
匹配关键词:
  - "model fit"
  - "performance"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 10
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "资料为arXiv上天体物理系外行星探测论文，与超节点/AI Rack/机柜级AI基础设施完全无关，未涉及任何核心主题。"
AI质检模型: "ali-deepseek-v4-flash"
AI质检时间: "2026-08-20T15:59:30+08:00"
AI主题相关性: 0
AI来源权威性: 5
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 5
AI摘要: "研究提出Astro-Hunters，用梯度提升分类器处理TESS两分钟测光数据检测系外行星凌星。实验发现标签来源对性能影响远大于模型架构，跨标签来源精确率-召回率差异达29倍，正确标注时AUC为0.788。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-07T03:24:38.218Z"
采集批次: "2026年8月20日14点19分32秒"
采集批次ID: "20260820-141932-079"
去重键: "https://arxiv.org/abs/2608.18172"
---

## Astrophysics > Earth and Planetary Astrophysics

## Title:Astro-Hunters: Machine Learning for Exoplanet Transit Detection in TESS Photometry

Authors:[Fatimah Emad Eldin](https://arxiv.org/search/astro-ph?searchtype=author&query=Eldin,+F+E)

[View PDF](https://arxiv.org/pdf/2608.18172) [HTML (experimental)](https://arxiv.org/html/2608.18172v1)

> Abstract:Finding planets beyond the Solar System is now largely a data-analysis problem: photometric surveys return far more stellar light curves than can be inspected by eye, and machine learning is increasingly asked to recognise the faint, periodic dimming of a transiting planet. Whether such a detector works turns on a choice that is seldom reported: how its training labels were made. No catalogue disposes the individual measurements a per-cadence detector must classify, so the annotation has to be derived, and this paper asks what that costs. We present Astro-Hunters, an end-to-end pipeline over TESS two-minute photometry: retrieval, detrending, seven sliding-window statistics per cadence, and gradient-boosted classification. These components are deliberately conventional. What is new is the treatment of label provenance as an experimental variable, a corpus of 189,279 cadences from twelve confirmed hosts annotated from published ephemerides rather than from the photometry, and a physical bound on what the task permits. Six classifier families are compared under a star-disjoint protocol. Holding features, model and protocol fixed, precision--recall performance spans a factor of 29 across label sources against 1.8 across architectures. Labels from an isolation forest fitted to the classifier's own features give an apparent AUC of 0.9915 that measures circularity; an unconverted transit epoch drives performance to chance; correct annotation gives AUC 0.788 at 5.3 times the prevalence baseline. The ceiling is observational, not architectural: a median single-cadence signal-to-noise ratio of 2.10 caps per-cadence AUC at 0.932. A Box Least Squares baseline recovers eight of twelve orbital periods from one sector. Phase-folding, not classifier capacity, is what makes the transit signal accessible.

| Subjects: | Earth and Planetary Astrophysics (astro-ph.EP); Instrumentation and Methods for Astrophysics (astro-ph.IM) |
| --- | --- |
| Cite as: | [arXiv:2608.18172](https://arxiv.org/abs/2608.18172) \[astro-ph.EP\] |
|  | (or [arXiv:2608.18172v1](https://arxiv.org/abs/2608.18172v1) \[astro-ph.EP\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2608.18172](https://doi.org/10.48550/arXiv.2608.18172) |

## Submission history

From: Fatimah Emad Eldin \[[view email](https://arxiv.org/show-email/8e9f3c11/2608.18172)\]  
**\[v1\]** Mon, 17 Aug 2026 18:41:30 UTC (205 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2608.18172) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))