The increasing popularity of long Text-to-Image (T2I) generation has created an urgent need for automatic and interpretable models that can evaluate the image-text alignment in long prompt scenarios. However, the existing T2I alignment benchmarks predominantly focus on short prompt scenarios and only provide MOS or Likert scale annotations. This inherent limitation hinders the development of long T2I evaluators, particularly in terms of the interpretability of alignment. In this study, we contribute LongT2IBench, which comprises 14K long text-image pairs accompanied by graph-structured human annotations. Given the detail-intensive nature of long prompts, we first design a Generate-Refine-Qualify annotation protocol to convert them into textual graph structures that encompass entities, attributes, and relations. Through this transformation, fine-grained alignment annotations are achieved based on these granular elements. Finally, the graph-structed annotations are converted into alignment scores and interpretations to facilitate the design of T2I evaluation models. Based on LongT2IBench, we further propose LongT2IExpert, a LongT2I evaluator that enables multi-modal large language models (MLLMs) to provide both quantitative scores and structured interpretations through an instruction-tuning process with Hierarchical Alignment Chain-of-Thought (CoT). Extensive experiments and comparisons demonstrate the superiority of the proposed LongT2IExpert in alignment evaluation and interpretation.

This study develops a machine learning (ML)-based multilayer cloud detection algorithm for the passive Advanced Himawari Imager (AHI) aboard the geostationary Himawari-8 satellite. AHI measurements in the 0.64-, 1.6-, 2.3-, 3.9-, 7.3-, 8.6-, 11.2-, and 12.4-$\mu \text{m}$channels, their combinations, geolocations, and observational geometries are used as predictors, and collocated active CloudSat and CALIPSO data are used to accurately label multilayer cloud pixels as the reference/truth of the predictand. We develop an ML-based daytime model (ML-Day) that utilizes all the aforementioned predictors and an all-time one (ML-All) that excludes the solar channel-dependent variables. Among four ML algorithms, the random forest (RF) performs slightly better than the artificial neural network, K-nearest neighbor, and support vector machines. By comparing with the merged CloudSat and CALIPSO product, the ML-Day model correctly identifies ~89% single-layer clouds and ~70% multilayer clouds, outperforming the Moderate Resolution Imaging Spectroradiometer (MODIS) operational multilayer cloud product (~80% and ~40% given by Marchantet al.). The success rates of ML-All for single-layer and multilayer clouds also reach ~85% and ~64%, respectively. The misclassification of our algorithm is mostly caused by missing optically thin clouds, a drawback of most radiometers without the 1.38-$\mu \text{m}$channel. Furthermore, with multilayer cloud pixels well detected by our algorithm, the AHI operational cloud top height retrievals are found to be larger biased due to multilayer cloud occurrence and might be improved by considering cloud vertical structures.

Near-infrared (NIR) channels, such as the 1.6- and 2.13-μm channels of Moderate Resolution Imaging Spectroradiometer (MODIS), play an important role in inferring cloud properties because of their sensitivity to cloud amount and particle size. Instead of the 2.13-μm channel, which has shown great success on MODIS, the central wavelength of the Visible Infrared Imaging Radiometer Suite (VIIRS) is shifted to 2.25 μm. This paper investigates the influences of NIR channels (i.e., 2.13 and $2.25 μm) on cloud optical and microphysical property retrievals and reveals the potential applications of the 2.25-μm channel to cloud thermodynamic phase and multilayer cloud detections by combining with the 1.6-μm channel. Rigorous radiative transfer simulations are performed to provide theoretical reflectance at the channels of interest, and MODIS and VIIRS observations are used for case studies. Our results indicate a minor influence of the 2.25-μm channel on cloud optical depth and effective particle size retrievals. In combination with the 1.6-μm channel, the 2.25-μm channel provides additional information indicating cloud phases. However, the 1.6- and 2.13-μm channels do not show any sensitivity to cloud phase. Furthermore, by considering the infrared-based cloud phase results, the 1.6- and 2.25-μm channel combination becomes possible to infer multilayer clouds. Case studies based on simultaneous MODIS and VIIRS observations demonstrate the capability of the 1.6-2.25-μm channel combination for determining cloud phase and multilayer clouds. Collocated satellite-based active lidar observations further validate these advantages of the 2.25-μmu channel over the original 2.13-μm channel.