technocracy

arXiv:2602.23478v1 Announce Type: new Abstract: Control Barrier Functions (CBFs) are a powerful tool for ensuring robotic safety, but designing or learning valid CBFs for complex systems is a significant challenge. While Hamilton-Jacobi Reachability provides a formal method for synthesizing safe value functions, it scales poorly and is typically performed offline, limiting its applicability in dynamic environments. This paper bridges the gap between offline synthesis and online adaptation. We introduce refineCBF for refining an approximate CBF – whether analytically […]

Ver mais

Fairness-aware machine learning has recently attracted various communities to mitigate discrimination against certain societal groups in data-driven tasks. For fair supervised learning, particularly in pre-processing, there have been two main categories: data fairness and task-tailored fairness. The former directly finds an intermediate distribution among the groups, independent of the type of the downstream model, so a learned downstream classification/regression model returns similar predictive scores to individuals inputting the same covariates irrespective of their sensitive attributes. The latter explicitly […]

Ver mais

arXiv:2505.09572v3 Announce Type: replace-cross Abstract: We study gradient flows for loss landscapes of fully connected feedforward neural networks with commonly used continuously differentiable activation functions such as the logistic, hyperbolic tangent, softplus or GELU function. We prove that the gradient flow either converges to a critical point or diverges to infinity while the loss converges to an asymptotic critical value. Moreover, we prove the existence of a threshold $varepsilon>0$ such that the loss value of any gradient flow […]

Ver mais

Web3 used to sound like a revolution. It was going to replace banks, fix the internet, destroy middlemen, return power to users, and rebuild digital life from the ground up. Every pitch felt massive. Every white paper sounded historic. Every project claimed it was creating the future. That energy helped Web3 grow. It also made Web3 hard to trust. Because somewhere along the way, the space became better at selling belief than shipping products. And that is why […]

Ver mais

arXiv:2605.00826v1 Announce Type: new Abstract: Text-to-video retrieval enables users to find relevant video content using natural language queries, a task that has grown increasingly important with the rapid expansion of online video. Over the past six years, research has produced numerous methods, such as dual encoders, attention-driven models, and multimodal fusion approaches; however, fundamental questions remain about model behavior, dataset influence, and query difficulty. In this work, we evaluate 14 state-of-the-art retrieval methods across 3 widely used datasets […]

Ver mais

This announcement is huge for the tech industry. Growth isn’t just about adding more people, it’s about how efficiently companies can work and what kind of technology they use. Meta’s decision to launch another round of layoffs, despite reporting strong financial results, shows there’s a bigger transformation happening across the sector. AI is taking center stage as the backbone of future operations. Just a few years back, Meta went through its “year of efficiency” and cut tens of […]

Ver mais

We develop a reinforcement learning algorithm to study the holographic entropy cone. Given a target entropy vector, our algorithm searches for a graph realization whose min-cut entropies match the target vector. If the target vector does not admit such a graph realization, it must lie outside the cone, in which case the algorithm finds a graph whose corresponding entropy vector most nearly approximates the target and allows us to probe the location of the facets. For the $sf […]

Ver mais

Multimodal Continual Instruction Tuning aims to continually enhance Large Vision Language Models (LVLMs) by learning from new data without forgetting previously acquired knowledge. Mixture of Experts (MoE) architectures naturally facilitate this by incrementally adding new experts and expanding routers while keeping the existing ones frozen. However, despite expert isolation, MoE-based continual learners still suffer from forgetting due to routing-drift: old-task tokens become mistakenly attracted to newly added experts, degrading performance on prior tasks. We analyze the failure mode […]

Ver mais

arXiv:2505.21441v4 Announce Type: replace Abstract: We propose a principled method for autoencoding with random forests. Our strategy builds on foundational results from nonparametric statistics and spectral graph theory to learn a low-dimensional embedding of the model that optimally represents relationships in the data. We provide exact and approximate solutions to the decoding problem via constrained optimization, split relabeling, and nearest neighbors regression. These methods effectively invert the compression pipeline, establishing a map from the embedding space back to […]

Ver mais