Been reading through “Masked Depth Modeling for Spatial Perception” from Ant Group and the core idea clicked for me. RGB-D cameras fail on reflective and transparent surfaces, and most methods just discard these missing values as noise. This paper does the opposite: sensor failures happen exactly where geometry is hardest (specular reflections, glass, textureless walls), so why not use them as natural masks for self-supervised learning? The setup takes full RGB as context, masks depth tokens where the […]

Ver mais

arXiv:2603.02376v1 Announce Type: new Abstract: Custom CUDA kernel development is essential for maximizing GPU utilization in large-scale distributed LLM training and inference, yet manually writing kernels that jointly leverage both computation and communication remains a labor-intensive and error-prone process. Prior work on kernel optimization has focused almost exclusively on computation, leaving communication kernels largely untouched even though they constitute a significant share of total execution time. We introduce CUCo, a training-free agent-driven workflow that automatically generates high-performance CUDA […]

Ver mais

arXiv:2602.21408v1 Announce Type: cross Abstract: Gaussian process (GP) surrogates are the default tool for emulating expensive computer experiments, but cubic cost, stationarity assumptions, and Gaussian predictive distributions limit their reach. We propose Generative Bayesian Computation (GBC) via Implicit Quantile Networks (IQNs) as a surrogate framework that targets all three limitations. GBC learns the full conditional quantile function from input–output pairs; at test time, a single forward pass per quantile level produces draws from the predictive distribution. Across fourteen […]

Ver mais

Ransomware core capability, unauthorized encryption, demands controls that identify and block malicious cryptographic activity without disrupting legitimate use. We present a probabilistic, risk-based access control architecture that couples machine learning inference with mandatory access control to regulate encryption on Linux in real time. The system builds a specialized dataset from the native ftrace framework using the function_graph tracer, yielding high-resolution kernel-function execution traces augmented with resource and I/O counters. These traces support both a supervised classifier and interpretable […]

Ver mais

Implementations of symmetric positive definite (SPD) matrix-based neural networks for neural decoding remain fragmented across research codebases and Python packages. Existing implementations often employ ad hoc handling of manifold constraints and non-unified training setups, which hinders reproducibility and integration into modern deep-learning workflows. To address this gap, we introduce SPD Learn, a unified and modular Python package for geometric deep learning with SPD matrices. SPD Learn provides core SPD operators and neural-network layers, including numerically stable spectral operators, […]

Ver mais

arXiv:2602.15834v1 Announce Type: new Abstract: We introduce a unified framework that combines nonlinear dynamics, perceptual psychophysics and high frequency haptic rendering to enhance realism in surgical simulation. The interaction of the surgical device with soft tissue is elevated to an augmented state space with a Koopman operator formulation, allowing linear prediction and control of the dynamics that are nonlinear by nature. To make the rendered forces consistent with human perceptual limits, we put forward a Bayesian calibration module […]

Ver mais

Ads are inevitable when you are on the internet. These days, Google Search is filled with sponsored results. Although AI has taken over our lives and how we browse the internet, ads are not leaving our backs. Earlier this month, OpenAI revealed its plans to display ads for users in the US on ChatGPT Free and Go plans. For the uninitiated, ChatGPT Go is now also live in the US, after initially debuting in the Indian and Indonesian […]

Ver mais

arXiv:2601.21868v1 Announce Type: new Abstract: Understanding the stability and long-time behavior of generative models is a fundamental problem in modern machine learning. This paper provides quantitative bounds on the sampling error of score-based generative models by leveraging stability and forgetting properties of the Markov chain associated with the reverse-time dynamics. Under weak assumptions, we provide the two structural properties to ensure the propagation of initialization and discretization errors of the backward process: a Lyapunov drift condition and a […]

Ver mais

arXiv:2601.06730v2 Announce Type: replace-cross Abstract: The Rashomon effect — the existence of multiple, distinct models that achieve nearly equivalent predictive performance — has emerged as a fundamental phenomenon in modern machine learning and statistics. In this paper, we explore the causes underlying the Rashomon effect, organizing them into three categories: statistical sources arising from finite samples and noise in the data-generating process; structural sources arising from non-convexity of optimization objectives and unobserved variables that create fundamental non-identifiability; and […]

Ver mais