arXiv:2511.10915v1 Announce Type: new Abstract: Federated clustering addresses the critical challenge of extracting patterns from decentralized, unlabeled data. However, it is hampered by the flaw that current approaches are forced to accept a compromise between performance and privacy: textit{transmitting embedding…
arXiv:2511.08544v3 Announce Type: replace Abstract: Learning manipulable representations of the world and its dynamics is central to AI. Joint-Embedding Predictive Architectures (JEPAs) offer a promising blueprint, but lack of practical guidance and theory has led to ad-hoc R&D. We present…
arXiv:2511.10936v1 Announce Type: new Abstract: Graph unlearning has emerged as a promising solution for complying with “the right to be forgotten” regulations by enabling the removal of sensitive information upon request. However, this solution is not foolproof. The involvement of…
arXiv:2305.06315v2 Announce Type: replace-cross Abstract: For deep learning problems on graph-structured data, pooling layers are important for down sampling, reducing computational cost, and to minimize overfitting. We define a pooling layer, nervePool, for data structured as simplicial complexes, which are…
arXiv:2511.10938v1 Announce Type: new Abstract: Motivated by the challenges of edge inference, we study a variant of the cascade bandit model in which each arm corresponds to an inference model with an associated accuracy and error probability. We analyse four…
arXiv:2511.07885v2 Announce Type: replace-cross Abstract: Large language model (LLM) queries are predominantly processed by frontier models in centralized cloud infrastructure. Rapidly growing demand strains this paradigm, and cloud providers struggle to scale infrastructure at pace. Two advances enable us to…
arXiv:2511.10572v2 Announce Type: replace-cross Abstract: Equitably allocating limited resources in high-stakes domains-such as education, employment, and healthcare-requires balancing short-term utility with long-term impact, while accounting for delayed outcomes, hidden heterogeneity, and ethical constraints. However, most learning-based allocation frameworks either assume…
arXiv:2502.13961v4 Announce Type: replace-cross Abstract: Understanding the advantages of deep neural networks trained by gradient descent (GD) compared to shallow models remains an open theoretical challenge. In this paper, we introduce a class of target functions (single and multi-index Gaussian…
arXiv:2508.02583v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have demonstrated strong performance across a wide range of tasks, yet they still struggle with complex mathematical reasoning, a challenge fundamentally rooted in deep structural dependencies. To address this challenge, we…
arXiv:2511.09392v3 Announce Type: replace Abstract: Sequential Recommenders, which exploit dynamic user intents through interaction sequences, is vulnerable to adversarial attacks. While existing attacks primarily rely on data poisoning, they require large-scale user access or fake profiles thus lacking practicality. In…
