The goal in thinning is to summarize a dataset using a small set of representative points. Remarkably, sub-Gaussian thinning algorithms like Kernel Halving and Compress can match the quality of uniform subsampling while substantially reducing the number of summary points. However, existing guarantees cover only a restricted range of distributions and kernel-based quality measures and suffer from pessimistic dimension dependence. To address these deficiencies, we introduce a new low-rank analysis of sub-Gaussian thinning that applies to any distribution and any kernel, guaranteeing high-quality compression whenever the kernel or data matrix is approximately low-rank. To demonstrate the broad applicability of the techniques, we design practical sub-Gaussian thinning approaches that improve upon the best known guarantees for approximating attention in transformers, accelerating stochastic gradient training through reordering, and distinguishing distributions in near-linear time.

Although modern neural networks often generalize to new combinations of familiar concepts, the conditions that enable such compositionality have long been an open question.In this work, we study the systematicity gap in visual question answering: the performance difference between reasoning on previously seen and unseen combinations of object attributes.To test, we introduce a novel diagnostic dataset, CLEVR-HOPE.We find that the systematicity gap is not reduced by increasing the quantity of training data, but is reduced by increasing the diversity of training data.In particular, our experiments suggest that the more distinct attribute type combinations are seen during training, the more systematic we can expect the resulting model to be.We release our data and code at https://github.com/ ikb-a/systematicity-gap-in-vqa.

CSCW researchers have long inquired into the ways that identity informs, and is informed by, the design of technological systems. Gender is a regularly considered aspect of identity, with extensive work documenting and exploring gendered experiences and designs with the aim of addressing inequalities in, or through, design. Recent work has questioned the way that we conceptualise and "measure" gender, advocating more nuanced classificatory schemes to avoid silencing or obscuring trans and/or non-binary experiences. Building on and extending this research, our work examines how gender is conceptualised more broadly. Drawing from a range of theoretical perspectives in gender studies, feminist and postcolonial theory, we argue for the treatment of gender as "multiplicitous" when we conceptualise and interpret research, in order to avoid unintentionally perpetuating silencing and inequality even as we work to tackle it. Illustrating our argument with examples from both within and without CSCW, we suggest both new research directions for CSCW scholars inquiring into gender, and sensitising questions that scholars can use when constructing and evaluating studies.