Search for: All records

Accurate identification of inundated areas is crucial for mitigating the impacts of flooding, which causes numerous casualties and significant economic losses. While polarimetric synthetic aperture radar (PolSAR) data have been utilized to detect inundated regions, the information contained within PolSAR features remains severely underutilized. We introduce a novel approach that involves extracting a large number of PolSAR features through various PolSAR decomposition techniques, selecting the most important ones using the decision tree–recursive feature elimination (DT-RFE) method, and ultimately detecting inundation using a convolutional neural network (CNN) model. The hybrid DT-RFE–CNN model was trained and tested over a region in southeastern North Carolina during Hurricane Florence on September 18, 2018, using PolSAR features derived from the Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR). In terms of flood-mapping efficacy, the DT-RFE–CNN model outperformed a CNN model that used only PolSAR data across all metrics in both the training and testing stages. The performance of the trained DT-RFE–CNN model was evaluated by testing it over the same region for four more days (September 19, 20, 22, and 23, 2018); it achieved an average accuracy, precision, recall, F1 score, and intersection-over-union of 0.9304, 0.9089, 0.9584, 0.9324, and 0.8738, respectively, outperforming both the classical Otsu method and the FT-Transformer model using features selected by DT-RFE. Finally, we assessed the model’s generalizability by mapping another significant flood event, caused by Hurricane Harvey in Texas between August and September 2017. Based on the results, the hybrid model can accurately detect flooding, even in regions on which it has not been trained. Thus, the proposed method can facilitate flood monitoring and response efforts. 

Moisture plays a key role in the energetics of hurricanes. Using a convolutional autoencoder, a state-of-the-art deep learning approach to spatial pattern classification, with k-means we identified four representative clusters of total column water vapor (TCWV) patterns around North Atlantic hurricanes. These four clusters exhibit distinct spatial distributions of TCWV in terms of amount, symmetry, and areal extent. Cluster 1 has a compact, symmetric, and moderate moisture pattern which we refer to as medium moisture symmetrical. Cluster 2 is high moisture symmetrical as these hurricanes have an abundance of moisture with a widespread and symmetric pattern. Cluster 3 is low moisture asymmetrical as it represents the driest conditions especially in the northwest. Cluster 4 has high moisture near the center but exhibits a pattern with the strongest contrast between dryness in the northwest and wetness in the southeast, thus we label it high moisture asymmetrical. Each cluster has distinct geographical and temporal distributions, indicating differences in dynamic and thermodynamic environmental conditions associated with each cluster's moisture pattern. Additionally, hurricane intensity, size, and precipitation features vary among the four clusters, characteristics which are closely associated with the moisture and environmental conditions of each cluster. Our study's application of a deep learning method in classifying spatial patterns of moisture around hurricanes highlights the importance of moisture conditions in a hurricane's evolution. 

Finetuned large language models (LLMs) have shown remarkable performance in financial tasks, such as sentiment analysis and information retrieval. Due to privacy concerns, finetuning and deploying financial LLMs (FinLLMs) locally are crucial for institutions and individuals. In this paper, we employ quantized low-rank adaptation (QLoRA) to finetune FinLLMs, which leverage low-rank structure and quantization technique to significantly reduce computational requirements while maintaining model performance. We also employ data and pipeline parallelism to enable local finetuning on commodity GPUs. Experiments on financial datasets validate the efficacy of our approach in yielding notable improvements over the base models. 

Financial large language models (FinLLMs) have been applied to various tasks in business, finance, accounting, and auditing. Complex financial regulations and standards are critical to financial services, which LLMs must comply with. However, FinLLMs’ performance in understanding and interpreting financial regulations has rarely been studied. Therefore, we organize the Regulations Challenge, a shared task at COLING FinNLP-FNP-LLMFinLegal2025. It encourages the academic community to explore the strengths and limitations of popular LLMs. We create 9 novel tasks and corresponding question sets. In this paper, we provide an overview of these tasks and summarize participants’ approaches and results. We aim to raise awareness of FinLLMs’ professional capability in financial regulations.