arXiv:2603.26781v1 Announce Type: new Abstract: With the rapid advancement of AI technology, we have seen more and more concerns on data privacy, leading to some cutting-edge research on machine learning with encrypted computation. Fully Homomorphic Encryption (FHE) is a crucial technology for privacy-preserving computation, while it struggles with continuous non-polynomial functions, as it operates on discrete integers and supports only addition and multiplication. Spiking Neural Networks (SNNs), which use discrete spike signals, naturally complement FHE’s characteristics. In this […]

Ver mais

arXiv:2505.18269v4 Announce Type: replace-cross Abstract: We study the problem of selecting a subset from a large action space shared by a family of bandits, with the goal of achieving performance nearly matching that of using the full action space. Indeed, in many natural situations, while the nominal set of actions may be large, there also exist significant correlations between the rewards of different actions. In this paper we propose an algorithm that can significantly reduce the action space […]

Ver mais

arXiv:2603.26696v1 Announce Type: new Abstract: In extreme environments such as underwater exploration and post-disaster rescue, tethered robots require continuous navigation while avoiding cable entanglement. Traditional planners struggle in these lifelong planning scenarios due to topological unawareness, while topology-augmented graph-search methods face computational bottlenecks in obstacle-rich environments where the number of candidate topological classes increases. To address these challenges, we propose Topological Motion Planning Diffusion (TMPD), a novel generative planning framework that integrates lifelong topological memory. Instead of relying […]

Ver mais

arXiv:2504.14772v2 Announce Type: replace-cross Abstract: The exponential growth of Large Language Models (LLMs) continues to highlight the need for efficient strategies to meet ever-expanding computational and data demands. This survey provides a comprehensive analysis of two complementary paradigms: Knowledge Distillation (KD) and Dataset Distillation (DD), both aimed at compressing LLMs while preserving their advanced reasoning capabilities and linguistic diversity. We first examine key methodologies in KD, such as task-specific alignment, rationale-based training, and multi-teacher frameworks, alongside DD techniques […]

Ver mais

Reinforcement Learning with Verifiable Rewards (RLVR) has become the leading paradigm for enhancing reasoning in Large Language Models (LLMs). However, standard RLVR algorithms suffer from a well-documented pathology: while they improve Pass@1 accuracy through sharpened sampling, they simultaneously narrow the model’s reasoning boundary and reduce generation diversity. We identify a root cause that existing methods overlook: the uniform penalization of errors. Current approaches — whether data-filtering methods that select prompts by difficulty, or advantage normalization schemes — treat […]

Ver mais

arXiv:2408.01517v3 Announce Type: replace-cross Abstract: We prove that the standard gradient flow in parameter space that underlies many training algorithms in deep learning can be continuously deformed into an adapted gradient flow which yields (constrained) Euclidean gradient flow in output space. Moreover, for the $L^{2}$ loss, if the Jacobian of the outputs with respect to the parameters is full rank (for fixed training data), then the time variable can be reparametrized so that the resulting flow is simply […]

Ver mais