arXiv:2603.04007v1 Announce Type: cross Abstract: We study fixed budget constrained best-arm identification in grouped bandits, where each arm consists of multiple independent attributes with stochastic rewards. An arm is considered feasible only if all its attributes’ means are above a given threshold. The aim is to find the feasible arm with the largest overall mean. We first derive a lower bound on the error probability for any algorithm on this setting. We then propose Feasibility Constrained Successive Rejects […]

Ver mais

arXiv:2602.10231v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) assigns a single scalar advantage to all tokens in a completion. For structured generations with explicit segments and objectives, this couples unrelated reward signals across segments, leading to objective interference and misattributed credit. We propose Blockwise Advantage Estimation, a family of GRPO-compatible methods that assigns each objective its own advantage and applies it only to the tokens in the corresponding text block, reducing reliance on hand-designed scalar rewards […]

Ver mais

arXiv:2601.18886v1 Announce Type: new Abstract: This paper introduces XProvence, a multilingual zero-cost context pruning model for retrieval-augmented generation (RAG), trained on 16 languages and supporting 100+ languages through effective cross-lingual transfer. Motivated by the growing use of RAG systems across diverse languages, we explore several strategies to generalize the Provence framework-which first integrated efficient zero-cost context pruning directly into the re-ranking model-beyond English. Across four multilingual question answering benchmarks, we show how XProvence can prune RAG contexts with […]

Ver mais

arXiv:2601.17005v1 Announce Type: new Abstract: The reliability of survey data is crucial in supply chain decision-making, particularly when evaluating readiness for AI-driven tools such as safety stock optimization systems. However, surveys often attract low-effort or fake responses that degrade the accuracy of derived insights. This study proposes a lightweight AI-based framework for filtering unreliable survey inputs using a supervised machine learning approach. In this expanded study, a larger dataset of 99 industry responses was collected, with manual labeling […]

Ver mais

arXiv:2601.04275v1 Announce Type: new Abstract: Machine unlearning aims to selectively remove the influence of specific training samples to satisfy privacy regulations such as the GDPR’s ‘Right to be Forgotten’. However, many existing methods require access to the data being removed, exposing it to membership inference attacks and potential misuse of Personally Identifiable Information (PII). We address this critical challenge by proposing Shadow Unlearning, a novel paradigm of approximate unlearning, that performs machine unlearning on anonymized forget data without […]

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

Hey everyone, I’ve been working on an RL project where I trained a PPO agent to control 4 elevators in a 20-floor building simulation. The goal was to see if RL can beat a classical Destination Dispatching algorithm. Results after 5M training steps on CPU: Classic agent: mean reward -0.67, avg wait 601 steps PPO agent: mean reward +0.14, avg wait 93 steps (~84% reduction) The hardest part was reward engineering – took several iterations to get dense […]

Ver mais

arXiv:2510.09895v3 Announce Type: replace-cross Abstract: Modeling clinical time-series data is hampered by the challenge of capturing latent, time-varying dependencies among features. State-of-the-art approaches often rely on black-box mechanisms or simple aggregation, failing to explicitly model how the influence of one clinical variable propagates through others over time. We propose $textbf{Chain-of-Influence (CoI)}$, an interpretable deep learning framework that constructs an explicit, time-unfolded graph of feature interactions. CoI enables the tracing of influence pathways, providing a granular audit trail that […]

Ver mais

arXiv:2603.06650v1 Announce Type: new Abstract: Whole-slide image classification for invasive lung adenocarcinoma subtyping remains vulnerable to real-world imaging perturbations that undermine model reliability at the decision boundary. We propose a margin consistency framework evaluated on 203,226 patches from 143 whole-slide images spanning five adenocarcinoma subtypes in the BMIRDS-LUAD dataset. By combining attention-weighted patch aggregation with margin-aware training, our approach achieves robust feature-logit space alignment measured by Kendall correlations of 0.88 during training and 0.64 during validation. Contrastive regularization, […]

Ver mais