Graph Neural Networks (GNNs) are increasingly being used for a variety of ML applications on graph data. Because graph data does not follow the independently and identically distributed (i.i.d.) assumption, adversarial manipulations or incorrect data can propagate to other data points through message passing, which deteriorates the model's performance. To allow model developers to remove the adverse effects of manipulated entities from a trained GNN, we study the recently formulated problem of Corrective Unlearning. We find that current graph unlearning methods fail to unlearn the effect of manipulations even when the whole manipulated set is known. We introduce a new graph unlearning method, Cognac, which can unlearn the effect of the manipulation set even when only 5% of it is identified. It recovers most of the performance of a strong oracle with fully corrected training data, even beating retraining from scratch without the deletion set while being 8x more efficient. We hope our work assists GNN developers in mitigating harmful effects caused by issues in real-world data post-training. Our code is publicly available at https://github.com/ Graph Neural Networks (GNNs) are seeing widespread adoption across diverse domains, from recommender systems to drug discovery (Wu et al., 2022; Zhang et al., 2022). Recently, GNNs are being scaled to large training sets for graph foundation models (Mao et al., 2024).

Language models have been shown to be rich enough to encode fMRI activations of certain Regions of Interest in our Brains. Previous works have explored transfer learning from representations learned for popular natural language processing tasks for predicting brain responses. In our work, we improve the performance of such encoders by creating an ensemble model out of 10 popular Language Models (2 syntactic and 8 semantic). We beat the current baselines by 10% on average across all ROIs through our ensembling methods.

Unsupervised Representation Learning on graphs is gaining gorithms on downstream tasks. Accordingly, fairness in the context traction due to the increasing abundance of unlabelled network of trained decision-making systems has increased in popularity recently data and the compactness, richness, and usefulness of the representations due to the numerous social distresses caused when systems generated. In this context, the need to consider fairness and that did not incorporate adequate fairness measures were deployed bias constraints while generating the representations has been wellmotivated in the wild [29, 25]. The job platform XING is an extreme example and studied to some extent in prior works. One major limitation that exhibited gender-based discrimination [4]. of most of the prior works in this setting is that they do not aim to address the bias generated due to connectivity patterns in the graphs, such as varied node centrality, which leads to a disproportionate performance across nodes. In our work, we aim to address this issue of mitigating bias due to inherent graph structure in an unsupervised setting. To this end, we propose CAFIN, a centralityaware fairness-inducing framework that leverages the structural information of graphs to tune the representations generated by existing frameworks.