Cool Embeddings:利用AstroCLIP预测IllustrisTNG和TNG-Cluster中的星系团冷却时间
Cool Embeddings: Predicting Galaxy Cluster Cooling Times in IllustrisTNG and TNG-Cluster with AstroCLIP
浏览论文内容
中文总结 AI 辅助
本研究利用AstroCLIP基础模型嵌入,从X射线图像直接预测星系团中心冷却时间,提出点积k近邻注意力模型,在TNG300和TNG-Cluster上实现低误差,并展现强泛化能力,为全天巡天中的冷核表征提供可扩展方案。
中文摘要 AI 辅助
基于光学天文数据训练的基础模型此前尚未被应用于X射线团热力学研究。我们展示了AstroCLIP(一个在光学星系图像和光谱上训练的对比视觉-语言模型)的嵌入在应用于模拟星系团的模拟钱德拉X射线图像时,保留了具有物理意义的结构——这是观测模态和物理尺度上的显著转变。我们针对星系团内介质的中心冷却时间($t_{\mathrm{cool},0}$)这一传统上需要X射线光谱学测量的关键热力学诊断指标,提出了一种直接从X射线图像预测$t_{\mathrm{cool},0}$的机器学习框架,并将其应用于来自IllustrisTNG模拟的TNG300中的711个投影和TNG-Cluster中的993个投影。AstroCLIP嵌入形成了一个结构化的潜在空间,其中具有相似热力学性质的星系团自然分组,而按晕质量则没有等效的分离。在此基础上,我们引入了一种点积$k$近邻注意力模型,该模型从训练集中检索相似星系团并自适应地重新加权其冷却时间。该模型在$\log(t_{\mathrm{cool},0})$上实现了0.24(TNG300)和0.23(TNG-Cluster)的均方根误差,并展现出强大的分布外泛化能力,优于在原始图像上训练的微调卷积神经网络(RMSE 0.50对比0.73)。这些结果表明,预训练基础模型的嵌入能够跨天体物理领域迁移,并编码了足以推断星系团冷却性质的热力学信息——为在eROSITA等全天巡天中实现冷核表征提供了一条可扩展的路径。
英文摘要
Foundation models trained on optical astronomical data have not previously been applied to X-ray cluster thermodynamics. We show that embeddings from AstroCLIP, a contrastive vision-language model trained on optical galaxy images and spectra, retain physically meaningful structure when applied to mock Chandra X-ray images of simulated galaxy clusters - a significant shift in observational modality and physical scale. We target the central cooling time ($t_{\mathrm{cool},0}$) of the intracluster medium, a key diagnostic of cluster thermodynamics traditionally requiring X-ray spectroscopy. We present a machine learning framework that predicts $t_{\mathrm{cool},0}$ directly from X-ray images, applied to 711 projections from TNG300 and 993 from TNG-Cluster from the IllustrisTNG simulations. AstroCLIP embeddings form a structured latent space in which clusters with similar thermodynamic properties are naturally grouped, with no equivalent separation by halo mass. Building on this, we introduce a dot-product $k$-nearest neighbour attention model that retrieves similar clusters from the training set and adaptively reweights their cooling times. This model achieves root-mean-square errors of 0.24 (TNG300) and 0.23 (TNG-Cluster) in $\log(t_{\mathrm{cool},0})$, and demonstrates strong out-of-distribution generalisation, outperforming a fine-tuned convolutional neural network trained on raw images (RMSE 0.50 versus 0.73). These results demonstrate that pretrained foundation model embeddings transfer across astrophysical domains and encode thermodynamic information sufficient for inference of cluster cooling properties - offering a scalable path toward cool-core characterisation in all-sky surveys such as eROSITA.
发表机构
- Université de Montréal(蒙特利尔大学)
- Centre de recherche en astrophysique du Québec (CRAQ)(魁北克天体物理学研究中心)
- Dragonfly Focused Research Organization(龙卷风重点研究组织)
- Max-Planck-Institut für Astronomie(马克斯·普朗克天文学研究所)
- University of Michigan(密歇根大学)
机构由 AI 辅助整理,请以论文原文为准。