Ira Assent

D-core decomposition (DCD) identifies groups of nodes with strong cohesion \dmsat different levels of granularity, supporting e.g. clustering and community detection in directed graphs. Given the inefficiency of computing DCD, the computational power of GPUs is an attractive target. However, existing algorithms are ill-suited due to their restrictive programming model, as well as memory constraints in the widely available consumer-level GPUs. In particular, we show that any trivial adaptation of peeling algorithms, state-of-the-art for DCD, suffers from workload imbalance, redundant operations, and repeated atomic operations. To address these challenges, we propose a novel dynamic strategy to balance the computation on nodes with varying vertex degree and an enumeration-based lightweight memory footprint with greatly reduced thread contention. We conduct extensive experiments on 12 datasets showing that Geld achieves an average speedup of 31× over GPU peeling DCD, 6× over GPU H-Index DCD, 22× over the best existing parallel DCD algorithm with 33% memory efficiency, and 3 orders of magnitude over the best serial DCD algorithm.

Large language models (LLMs) are increasingly used for everyday communication tasks, including drafting interpersonal messages intended to influence and persuade. Prior work has shown that LLMs can successfully persuade humans and amplify persuasive language. It is therefore essential to understand how user instructions affect the generation of persuasive language, and to understand whether the generated persuasive language differs, for example, when targeting different groups. In this work, we propose a framework for evaluating how persuasive language generation is affected by recipient gender, sender intent, or output language. We evaluate 13 LLMs and 16 languages using pairwise prompt instructions. We evaluate model responses on 19 categories of persuasive language using an LLM-as-judge setup grounded in social psychology and communication science. Our results reveal significant gender differences in the persuasive language generated across all models. These patterns reflect biases consistent with gender-stereotypical linguistic tendencies documented in social psychology and sociolinguistics.

Explanation for deep learning models on time series classification (TSC) tasks is an important and challenging problem. Most existing approaches use attribution maps to explain outcomes. However, they have limitations in generating explanations that are well-aligned with humans's perceptions. Recently LIME-based approaches provide a more meaningful explanation via segmenting the data. However, these approaches are still suffering from the processes of segment generations and evaluations. In this paper, we propose a novel time series explanation approach called InteDisUX to overcome these problems. Our technique utilizes the segment-level integrated gradient (SIG) for calculating importance scores for an initial set of small and equal segments before iteratively merge two consecutive ones to create better explanations under a unique greedy strategy guided by two new proposed metrics including discrimination and faithfulness gains. By this way, our method does not depend on predefined segments like others while being robusts to instability, poor local fidelity and data imbalance like LIME-based methods. Furthermore, InteDisUX is the first work to use the model's information to improve the set of segments} for time series explanation. Extensive experiments show that our method outperforms LIME-based ones in 12 datasets in terms of faithfulness and 8/12 datasets in terms of robustness.

Knowledge graphs form large networks of millions of entities (e.g., Michelle Obama, Barack Obama) and relationships (e.g., married). To obtain an overview of the entity, we need to inspect a potentially large number of relationships to other entities. For this reason, entity summarization aims to extract succinct but expressive descriptions of each entity. Yet, existing methods build their summaries only based on the immediate connections of an entity, disregarding how indirect relationships contain essential information for describing the entity (e.g., understanding Michelle Obama also via her husband's role as former president). We propose IRES, an unsupervised entity summarization method built on graph theoretical principles. We draw a notable connection between the informativeness of a summary and graph partitioning, and devise an effective approach to learn diverse aspects that characterize an entity. In a comprehensive experimental study, IRES shows superior summary quality. In particular, when full neighborhood information is available, IRES outperforms existing methods by 6 percentage points F1 while maintaining competitive computational efficiency.

We introduce MOSCITO (MOlecular Dynamics Subspace Clustering with Temporal Observance), a subspace clustering for molecular dynamics data. MOSCITO groups those timesteps of a molecular dynamics trajectory together into clusters in which the molecule has similar conformations. In contrast to state-of-the-art methods, MOSCITO takes advantage of sequential relationships found in time series data. Unlike existing work, MOSCITO does not need a two-step procedure with tedious post-processing, but directly models essential properties of the data. Interpreting clusters as Markov states allows us to evaluate the clustering performance based on the resulting Markov state models. In experiments on 60 trajectories and 4 different proteins, we show that the performance of MOSCITO achieves state-of-the-art performance in a novel single-step method. Moreover, by modeling temporal aspects, MOSCITO obtains better segmentation of trajectories, especially for small numbers of clusters.

This paper presents a solution to the Weather4Cast 2023 competition, where the goal is to forecast high-resolution precipitation with an 8-hour lead time using lower-resolution satellite radiance images. We propose a simple, yet effective method for spatiotemporal feature learning using a 2D U-Net model, that outperforms the official 3D U-Net baseline in both performance and efficiency. We place emphasis on refining the dataset, through importance sampling and dataset preparation, and show that such techniques have a significant impact on performance. We further study an alternative cross-entropy loss function that improves performance over the standard mean squared error loss, while also enabling models to produce probabilistic outputs. Additional techniques are explored regarding the generation of predictions at different lead times, specifically through Conditioning Lead Time. Lastly, to generate high-resolution forecasts, we evaluate standard and learned upsampling methods. The code and trained parameters are available at https://github.com/rafapablos/w4c23-rainai.