// radar de ia

Visão Computacional

Papers, modelos e datasets em alta no Hugging Face, além do blog oficial — com leitura editorial em português.

todas LLMs & Texto Geração de Imagem Visão Computacional Áudio & Voz Multimodal Dados & Embeddings Robótica & RL
Blog Geração de Imagem

Modeling Decisions in Blockchain Analytics: A Leakage-Aware Evaluation of Tree-Based vs. Sequential Models

arXiv:2607.27350v1 Announce Type: new Abstract: Sybil bots are Ethereum actors that imitate legitimate users to extract airdrop rewards or influence governance. Recent Sybil detection methods increasingly use deep learning and treat blockchain activity as a quasi-linguistic sequence. However, complex sequence models are computationally expensive for real-time monitoring, and their reported performance may be inflated by label leakage from high-signal smart contracts. We ask whether and how organ...

31.07.2026
Blog Visão Computacional

ZMIS-SAM: Segment Anything Model Enhanced with Wavelet Transform for Zooplankton Microscopy Image Instance Segmentation

arXiv:2607.27585v1 Announce Type: new Abstract: As primary consumers in the marine food chain, zooplankton play a crucial role in maintaining marine ecological balance. However, the Segment Anything Model (SAM) exhibits limited performance in microscopic image instance segmentation due to its lack of zooplankton-specific domain knowledge. To address these challenges, we propose a novel instance segmentation model based on SAM and wavelet transform (ZMIS-SAM), effectively tackling issues such as ...

31.07.2026
Blog Robótica & RL

How AI is Changing Linux VPS Security for Businesses

Cybersecurity wasn’t really something small and mid-sized businesses worried about too much a few years back. That’s changed fast. Once your customer data, your apps, your internal tools all end up online, protecting the server behind them stops being optional. And attacks aren’t getting any simpler either, which is part of the problem with sticking […] The post How AI is Changing Linux VPS Security for Businesses appeared first on AI News .

30.07.2026
Blog Geração de Imagem

Um framework baseado em GAN para síntese robusta de dados em observações de Internet via satélite

arXiv:2607.24790v1 Tipo de anúncio: novo Resumo: A Internet via satélite em órbita terrestre baixa (LEO) tornou-se uma infraestrutura importante para viabilizar conectividade onipresente, em consonância com a visão da União Internacional de Telecomunicações para as redes de telecomunicações 6G. No entanto, as observações atuais de Internet via satélite LEO frequentemente sofrem com dados ausentes, o que complica a tarefa de aumento de dados e limita a expansão de conjuntos de dados representativos. Dadas as características complexas desses conjuntos de dados, gen...

29.07.2026
Blog LLMs & Texto

GLIDE: Atenção Híbrida Guiada por Camada para Inferência Eficiente de LLM

arXiv:2607.24788v1 Tipo de anúncio: novo Resumo: À medida que os Large Language Models escalam para contextos cada vez mais longos, o overhead de I/O de memória e de computação do cache de Chave-Valor (KV) durante a decodificação surge como o principal gargalo de throughput. Para resolver isso, propomos o GLIDE, uma Atenção Híbrida Guiada por Camada que integra estrategicamente a atenção softmax de janela deslizante com a agregação recorrente linear. O GLIDE é motivado pela heterogeneidade entre camadas: as camadas iniciais exibem alta sensibilidade a ...

29.07.2026
Blog Visão Computacional

When Does Deep Representation Learning Help Single-Cell Clustering? A Sensitivity-Aware Diagnostic Benchmark for Biomedical AI Pipelines

arXiv:2607.25288v1 Announce Type: new Abstract: Single-cell ribonucleic acid sequencing (scRNA-seq) is a foundational technology for precision-medicine workflows that contribute to United Nations Sustainable Development Goal 3 on Good Health and Well-being, and unsupervised clustering is the analytical step that turns raw expression matrices into interpretable cell populations. Practitioners therefore face a recurring engineering decision: is an additional deep representation stage worth its com...

29.07.2026
Blog Visão Computacional

CD-RMOT-Bench: Benchmarking the Cross-Domain Referring Multi-Object Tracking

arXiv:2607.25239v1 Announce Type: new Abstract: Referring multi-object tracking (RMOT) extends tracking from category-driven perception to language-guided understanding by grounding object trajectories in natural-language expressions. Despite recent progress, existing RMOT studies are largely conducted under in-domain settings, leaving the robustness of language-conditioned tracking under inevitable visual domain shifts unexplored. In this paper, we study Cross-Domain Referring Multi-Object Trac...

29.07.2026
Blog Visão Computacional

Aletheia: Um Sistema de Apoio à Decisão Clínica que Prioriza o Funcionamento Offline para Diagnóstico Diferencial em Ambientes de Saúde com Poucos Recursos

arXiv:2607.24814v1 Tipo de anúncio: novo Resumo: O acesso à expertise clínica especializada continua severamente limitado em toda a África subsaariana, onde a proporção entre médicos e pacientes pode cair abaixo de 1:25.000 em áreas rurais. As ferramentas de diagnóstico assistidas por IA existentes exigem predominantemente conectividade de internet confiável e hardware de alta especificação, tornando-as impraticáveis para os profissionais de saúde da linha de frente em hospitais distritais e centros de saúde. Este artigo apresenta o Aletheia, um sistema de apoio à decisão clínica que prioriza o funcionamento offline...

29.07.2026
Blog LLMs & Texto

PILA: Plug-and-Play Insertion for LLM-native Advertising

arXiv:2607.25590v1 Announce Type: new Abstract: How to monetize large language models (LLMs) by naturally integrating sponsored content into their responses, known as LLM-native advertising, has recently emerged as a critical problem. However, existing solutions entangle advertising with content generation inside a single model, which is incompatible with modern API-only or workflow-based LLM applications and inevitably compromises the original response quality. To address this, we propose PILA,...

29.07.2026
← anterior 3 / 30 próxima →
350 itens no radar