Papers, modelos e datasets em alta no Hugging Face, além do blog oficial — com leitura editorial em português.
LaMem-VLA introduces a latent-memory-native framework that integrates historical experience into vision-language-action reasoning through coordinated memory components operating in…
SciReasoner is a multimodal scientific foundation model that enables interpretable structural reasoning across proteins, molecules, and crystals by discretizing structural elements…
General Intuition, Kyutai e Epic Games publicam sob Apache 2.0 um modelo de mundo de 5 bilhões de parâmetros que roda partidas 2v2 de Rocket League a 20 quadros por segundo numa única GPU — junto com 29 terabytes de dados de treino.
Modelo de geração de texto · 27 B de parâmetros — 4.5 mil downloads e 44 curtidas no Hugging Face.
arXiv:2607.02588v1 Announce Type: new Abstract: Multimodal large language models excel on short clips but struggle on hour-long videos in an online setting, where frames are processed incrementally under limited memory. Existing online methods either retain compact visual representations that lack semantic structure, or build higher-level memory stores organized around temporal proximity rather than explicit causal links, leaving multi-hop narrative reasoning to be reconstructed by the LLM at ev...
arXiv:2607.02568v1 Announce Type: new Abstract: View-based point cloud completion aims to recover a complete 3D shape from a partial point cloud, guided by a single-view image. However, existing approaches often suffer from limited performance due to weak modality alignment and limited self-geometry enhancement. To overcome these challenges, we propose a unified geometry-aware framework that integrates efficient modality alignment and adaptive geometry enhancement, mainly to address cross-modal ...
arXiv:2607.03146v1 Announce Type: new Abstract: Vision-language-action (VLA) models open a new path toward intuitive robot control by directly linking perception, language, and action in a single end-to-end framework. Yet for UAVs, practical adoption remains difficult because existing solutions are either computationally heavy or insufficiently capable in complex environments. In this work, we propose a practical expert-distillation pipeline (Exp2VLA) for language-conditioned drone navigation. T...
arXiv:2607.03387v1 Announce Type: new Abstract: Tactile perception is indispensable for contact-rich manipulation, yet integrating it into Vision-Language-Action (VLA) models often induces modality collapse, where high-bandwidth visual features overshadow sparse tactile cues. Inspired by Predictive Coding, a neural mechanism where the brain attenuates predictable inputs to prioritize surprising stimuli, we propose ResTacVLA. Rather than treating tactile data as raw input, we reformulate it as a ...
arXiv:2607.02585v1 Announce Type: new Abstract: Multimodal CT-MRI registration is central to image-guided radiotherapy, surgical navigation, and diagnostic workflows, but most pipelines report only aggregate quality metrics without per-case reliability signals. We propose a reliability-aware framework that converts registration quality into Green/Yellow/Red risk categories using data-learned thresholds. CT images were registered to T1-weighted MRI using rigid and affine transformations on 90 pai...
arXiv:2607.02592v1 Announce Type: new Abstract: On-policy distillation (OPD) has recently emerged as an effective post-training paradigm by providing supervision on student-generated trajectories. However, existing OPD methods for multimodal reasoning usually rely on a static teacher routing, assigning each sample to a single teacher based on modality or task type. This ignores that visual grounding and abstract reasoning may dominate different decoding steps, making a single teacher insufficien...
arXiv:2607.02862v1 Announce Type: new Abstract: Automatic Speech Recognition (ASR) and Dialect Identification (DID) are crucial for Indian languages, many of which are low-resource and exhibit significant dialectal differences. Existing methods often optimize ASR or DID individually, resulting in performance trade-offs. In this work, we propose a multimodal framework that jointly improves ASR and DID. Our method employs a Bottleneck Encoder to extract dialectal features from Conformer-based spee...
arXiv:2607.03530v1 Announce Type: new Abstract: We introduce MentalThink, a visual-symbolic reasoning paradigm that equips Multimodal LLMs (MLLMs) with an executable mechanism for "mental" visualization. The core of MentalThink is a think-with-SVG pipeline, where the model learns to generate, render, and interpret scalable vector graphics (SVG) code as an intermediate visual representation for multi-turn reasoning. By creating structured vector sketches, the model can externalize spatial hypothe...