Recent advancements in graph neural networks (GNNs) have highlighted the critical need of calibrating model predictions, with neighborhood prediction similarity recognized as a pivotal component. Existing studies suggest that nodes with analogous neighborhood prediction similarity often exhibit similar calibration characteristics. Building on this insight, recent approaches incorporate neighborhood similarity into node-wise temperature scaling techniques. However, our analysis reveals that this assumption does not hold universally. Calibration errors can differ significantly even among nodes with comparable neighborhood similarity, depending on their confidence levels. This necessitates a re-evaluation of existing GNN calibration methods, as a single, unified approach may lead to sub-optimal calibration. In response, we introduce Simi-Mailbox, a novel approach that categorizes nodes by both neighborhood similarity and their own confidence, irrespective of proximity or connectivity. Our method allows fine-grained calibration by employing group-specific temperature scaling, with each temperature tailored to address the specific miscalibration level of affiliated nodes, rather than adhering to a uniform trend based on neighborhood similarity. Extensive experiments demonstrate the effectiveness of our Simi-Mailbox across diverse datasets on different GNN architectures, achieving up to 13.79% error reduction compared to uncalibrated GNN predictions.

Predicting the ground-state 3D molecular conformations from 2D molecular graphs is critical in computational chemistry due to its profound impact on molecular properties. Deep learning (DL) approaches have recently emerged as promising alternatives to computationally-heavy classical methods such as density functional theory (DFT). However, we discover that existing DL methods inadequately model inter-atomic forces, particularly for non-bonded atomic pairs, due to their naive usage of bonds and pairwise distances. Consequently, significant prediction errors occur for atoms with low degree (ie., low coordination numbers) whose conformations are primarily influenced by non-bonded interactions. To address this, we propose ReBIND, a novel framework that rewires molecular graphs by adding edges based on the Lennard-Jones potential to capture non-bonded interactions for low-degree atoms. Experimental results demonstrate that ReBIND significantly outperforms state-of-the-art methods across various molecular sizes, achieving up to a 20% reduction in prediction error. The code is available in: https://github.com/holymollyhao/ReBIND

Since the pioneering work on the lottery ticket hypothesis for graph neural networks (GNNs) was proposed in Chen et al. (2021), the study on finding graph lottery tickets (GLT) has become one of the pivotal focus in the GNN community, inspiring researchers to discover sparser GLT while achieving comparable performance to original dense networks. In parallel, the graph structure has gained substantial attention as a crucial factor in GNN training dynamics, also elucidated by several recent studies. Despite this, contemporary studies on GLT, in general, have not fully exploited inherent pathways in the graph structure and identified tickets in an iterative manner, which is time-consuming and inefficient. To address these limitations, we introduce **TEDDY**, a one-shot edge sparsification framework that leverages structural information by incorporating *edge-degree* statistics. Following the edge sparsification, we encourage the parameter sparsity during training via simple projected gradient descent on the $\ell_0$ ball. Given the target sparsity levels for both the graph structure and the model parameters, our TEDDY facilitates efficient and rapid realization of GLT within a *single* training. Remarkably, our experimental results demonstrate that TEDDY significantly surpasses conventional iterative approaches in generalization, even when conducting one-shot sparsification that solely utilizes graph structures, without taking feature information into account.

Understanding point clouds captured from the real-world is challenging due to shifts in data distribution caused by varying object scales, sensor angles, and self-occlusion. Prior works have addressed this issue by combining recent learning principles such as self-supervised learning, self-training, and adversarial training, which leads to significant computational overhead. Toward succinct yet powerful domain adaptation for point clouds, we revisit the unique challenges of point cloud data under domain shift scenarios and discover the importance of the global geometry of source data and trends of target pseudo-labels biased to the source label distribution. Motivated by our observations, we propose an adapter-guided domain adaptation method, PC-Adapter, that preserves the global shape information of the source domain using an attention-based adapter, while learning the local characteristics of the target domain via another adapter equipped with graph convolution. Additionally, we propose a novel pseudo-labeling strategy resilient to the classifier bias by adjusting confidence scores using their class-wise confidence distributions to consider relative confidences. Our method demonstrates superiority over baselines on various domain shift settings in benchmark datasets - PointDA, GraspNetPC, and PointSegDA.

Studies suggest women dropout of college and leave the workforce due to their family, finances, and military duty. However, women interested in (re-)entering the tech fields can be the largest untapped talent pool that may fulfill the needs of the future computing workforce. In this panel, five passionate women will share their experiences with identifying the challenges for women to re-enter emerging technology professions and the role of industry-academic relationship in facilitating such initiatives in order to develop future relevant initiatives.

As society increasingly relies on digital technologies in many different aspects, those who lack relevant access and skills are lagging increasingly behind. Among the underserved groups disproportionately affected by the digital divide are women who are transitioning from incarceration and seeking to reenter the workforce outside the carceral system (women-in-transition). Women-in-transition rarely have been exposed to sound technology education, as they have generally been isolated from the digital environment while in incarceration. Furthermore, while women have become the fastest-growing segment of the incarcerated population in the United States in recent decades, prison education and reentry programs are still not well adjusted for them. Most programs are mainly designed for the dominant male population. Consequently, women-in-transition face significant post-incarceration challenges in accessing and using relevant digital technologies and thus have added difficulties in entering or reentering the workforce. Against this backdrop, our multi-disciplinary research team has conducted empirical research as part of technology education offered to women-in-transition in the Midwest. In this article, we report results from our interviews with 75 women-in-transition in the Midwest that were conducted to develop a tailored technology education program for the women. More than half of the participants in our study are women of color and face precarious housing and financial situations. Then, we discuss principles that we adopted in developing our education program for the marginalized women and participants’ feedback on the program. Our team launched in-person sessions with women-in-reentry at public libraries in February 2020 and had to move the sessions online in March due to COVID-19. Our research-informed educational program is designed primarily to support the women in enhancing their knowledge and comfort with technology and nurturing computational thinking. Our study shows that low self-efficacy and mental health challenges, as well as lack of resources for technology access and use, are some of the major issues that need to be addressed in supporting technology learning among women-in-transition. This research offers scholarly and practical implications for computing education for women-in-transition and other marginalized populations.

This article develops a preferential attachment‐based mixture model of global Internet bandwidth and investigates it in the context of observed bandwidth distributions between 2002 and 2011. Our longitudinal analysis shows, among other things, that the bandwidth share distributions—and thus bandwidth differences—exhibit considerable path dependence where country proportions of international bandwidth in 2011 can be substantially accounted for by a preferential attachment‐based mixture of micro‐level processes. Our preferential attachment model, consistent with empirical data, does not predict increasing concentration of bandwidth within top‐ranked countries. We argue that recognizing the strong, but nuanced, historical inertia of bandwidth distributions is helpful in better discriminating among competing theoretical perspectives on the global digital divide as well as in clarifying policy discussions related to gaps between bandwidth‐rich and bandwidth‐poor countries.