NITRO-D: Native Integer-only Training of Deep Convolutional Neural Networks

Large language models (LLMs) have shown significant advancements in code generation, yet disparities remain in performance across various programming languages. To bridge this gap, a new approach called On-Policy Optimization with Group Equivalent Preference Optimization (GEPO) has been introduced, leveraging code translation tasks and a novel reinforcement learning framework known as OORL.

Recent research highlights a paradox in deep learning, revealing that adversarially trained models, designed to enhance robustness against attacks, may inadvertently increase the transferability of adversarial examples. This study involved training 36 diverse models, including CNNs and ViTs, and conducting extensive transferability experiments, leading to significant findings about model vulnerabilities.

FeatureLens has been introduced as a lightweight framework designed to detect adversarial examples in image classification, addressing the vulnerabilities of deep neural networks (DNNs) to such attacks. The framework utilizes an Image Feature Extractor and shallow classifiers, achieving high detection accuracy across various adversarial attack methods while maintaining interpretability and generalization.

A recent study assessed the alignment of popular Convolutional Neural Networks (CNNs) with human brain processes related to valence appraisal, revealing that these models struggle to reflect higher-order cognitive functions beyond basic visual processing. The research utilized correlation analysis with human behavioral and fMRI data to evaluate this alignment.

A recent study published on arXiv presents a novel approach to modeling and classifying energy consumption profiles across large distributed territories using pretopology. This method aims to optimize building energy management by automating the recommendations system, thus reducing the need for extensive manual audits of thousands of buildings.

A recent study highlights the implementation of an AI framework within Shriners Childrens, focusing on enhancing data quality in their Research Data Warehouse. The modernization to OMOP CDM v5.4 and the introduction of a Python-based data quality assessment tool aim to address existing challenges in AI system evaluations and clinical adoption.

The ECHO-MIMIC framework has been introduced to address collective action problems by transforming ill-structured problems into executable heuristics. This two-stage process involves evolving Python code for behavioral policies and generating persuasive messages to encourage agent compliance with these policies.

A recent study has introduced a hierarchical attention model for detecting sparse volumetric anomalies in subclinical keratoconus using 3D anterior segment OCT volumes. This model was compared against sixteen modern deep learning architectures, revealing superior performance in sensitivity and specificity over traditional 2D/3D CNNs and ViTs.