Offline meta-reinforcement learning seeks to learn policies that generalize across related tasks from fixed datasets. Context-based methods infer a task representation from transition histories, but learning effective task representations without supervision remains a challenge. In parallel, latent world models have demonstrated strong self-supervised representation learning through temporal consistency. We introduce contextual latent world models, which condition latent world models on inferred task representations and train them jointly with the context encoder. This enforces task-conditioned temporal consistency, yielding task […]

Ver mais

arXiv:2603.10016v1 Announce Type: new Abstract: We investigate whether large language models (LLMs) display human-like cognitive biases, focusing on potential implications for assistance in judicial sentencing, a decision-making system where fairness is paramount. Two of the most relevant biases were chosen: the virtuous victim effect (VVE), with emphasis given to its reduction when adjacent consent is present, and prestige-based halo effects (occupation, company, and credentials). Using vignettes that were altered from prior literature to avoid LLMs recalling from their […]

Ver mais

Learning activation functions has emerged as a promising direction in deep learning, allowing networks to adapt activation mechanisms to task-specific demands. In this work, we introduce a novel framework that employs the Gumbel-Softmax trick to enable discrete yet differentiable selection among a predefined set of activation functions during training. Our method dynamically learns the optimal activation function independently of the input, thereby enhancing both predictive accuracy and architectural flexibility. Experiments on synthetic datasets show that our model consistently […]

Ver mais

arXiv:2602.05023v1 Announce Type: new Abstract: Vision-language models (VLMs) have demonstrated strong performance in image geolocation, a capability further sharpened by frontier multimodal large reasoning models (MLRMs). This poses a significant privacy risk, as these widely accessible models can be exploited to infer sensitive locations from casually shared photos, often at street-level precision, potentially surpassing the level of detail the sharer consented or intended to disclose. While recent work has proposed applying a blanket restriction on geolocation disclosure to […]

Ver mais

arXiv:2602.15072v1 Announce Type: new Abstract: Accurate polyp segmentation in colonoscopy is essential for cancer prevention but remains challenging due to: (1) high morphological variability (from flat to protruding lesions), (2) strong visual similarity to normal structures such as folds and vessels, and (3) the need for robust multi-scale detection. Existing deep learning approaches suffer from unidirectional processing, weak multi-scale fusion, and the absence of anatomical constraints, often leading to false positives (over-segmentation of normal structures) and false negatives […]

Ver mais

arXiv:2602.05017v1 Announce Type: new Abstract: Agent-based cellular models simulate tissue evolution by capturing the behavior of individual cells, their interactions with neighboring cells, and their responses to the surrounding microenvironment. An important challenge in the field is scaling cellular resolution models to real-scale tumor simulations, which is critical for the development of digital twin models of diseases and requires the use of High-Performance Computing (HPC) since every time step involves trillions of operations. We hereby present a scalable […]

Ver mais

Hebbian Learning in Neural Networks One major difference between human and machine learning is the way we retain important aspects of our knowledge, as we gather more data. All throughout our life, we keep enforcing those concepts/facts which help us most in our day-to-day activities. ML algorithms are much less selective – if you train a NN to perform task A and then retrain it to perform task B, the parameters for A will be forgotten ‘uniformly’. This […]

Ver mais

arXiv:2602.08026v1 Announce Type: cross Abstract: We analyse linear ensemble sampling (ES) with standard Gaussian perturbations in stochastic linear bandits. We show that for ensemble size $m=Theta(dlog n)$, ES attains $tilde O(d^{3/2}sqrt n)$ high-probability regret, closing the gap to the Thompson sampling benchmark while keeping computation comparable. The proof brings a new perspective on randomized exploration in linear bandits by reducing the analysis to a time-uniform exceedance problem for $m$ independent Brownian motions. Intriguingly, this continuous-time lens is not […]

Ver mais

arXiv:2604.07421v1 Announce Type: new Abstract: Full-waveform inversion (FWI) is pivotal for reconstructing high-resolution subsurface velocity models but remains computationally intensive and ill-posed. While deep learning approaches promise efficiency, existing Convolutional Neural Networks (CNNs) and single-paradigm Neural Operators (NOs) struggle with one fundamental issue: frequency entanglement of multi-scale geological features. To address this challenge, we propose Spectral-Preserving Adaptive MoE (SPAMoE), a novel spectrum-aware framework for solving inverse problems with complex multi-scale structures. Our approach introduces a Spectral-Preserving DINO Encoder […]

Ver mais