Multimodal — Radar de IA

Unleashing More Actions via Action Compositional Training for VLA Models

arXiv:2607.00351v1 Announce Type: new Abstract: Vision-Language-Action models excel at robotic manipulation, driven by the scale and diversity of demonstration data. However, standard training paradigms often cause VLA models to severely overfit to specific behavioral patterns, rendering them unable to generalize to out-of-distribution scenarios even when those scenarios merely require novel combinations of identical sub-skills. While expanding datasets can mitigate this overfitting, acquiring h...

02.07.2026

Blog Visão Computacional

From Technical Metrics to User Perception: A User Study of a Multimodal Human-Robot Interaction System for Object Detection and Grasping

arXiv:2607.00530v1 Announce Type: new Abstract: Improvements in the technical performance of human--robot interaction (HRI) systems do not automatically translate into differences that human users can detect during live interaction. This paper investigates whether a 15 percentage point gain in end-to-end task success (from 75% in a multimodal baseline system to 90% in an improved configuration identified through a prior ablation study) is sufficient to produce consistent and measurable differenc...

02.07.2026

Blog LLMs & Texto

EgoSafetyBench: A Diagnostic Egocentric Video Benchmark for Evaluating Embodied VLMs as Runtime Safety Guards

arXiv:2607.00218v1 Announce Type: new Abstract: Vision-language models (VLMs) are now proposed as runtime safety guards for embodied agents in homes and factories. A deployable guard must catch genuinely unsafe situations while avoiding unnecessary intervention on routine but superficially alarming activity, a distinction that binary safety benchmarks obscure. We introduce EgoSafetyBench, an egocentric video benchmark of 1,200 robot-view scenarios annotated at half-second granularity, to evaluat...

02.07.2026

Blog Multimodal

Learning When to Listen: Gated Affect Fusion for Human Motion Prediction

arXiv:2607.00296v1 Announce Type: new Abstract: Human motion forecasting in unconstrained real-world videos remains challenging due to the ambiguity of future behaviors and the presence of noisy multimodal observations. While facial affect potentially provides complementary behavioral cues, its practical utility and mechanistic boundaries within motion forecasting frameworks remain poorly understood. In this work, we present a systematic study investigating the utility and temporal limitations o...

02.07.2026

Paper LLMs & Texto

AnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models

Vision-Language Models struggle with domain adaptation in specialized spatio-temporal video grounding tasks, highlighting limitations in zero-shot generalization and in-context lea…

02.07.2026 ·▲ 8

Paper Robótica & RL

Embodied.cpp: A Portable Inference Runtime of Embodied AI Models on Heterogeneous Robots

Embodied.cpp is a portable C++ runtime that enables efficient deployment of vision-language-action and world-action models across heterogeneous edge devices through modular executi…

02.07.2026 ·▲ 15

Paper Multimodal

Learning to Move Before Learning to Do: Task-Agnostic pretraining for VLAs

Task-Agnostic Pretraining framework trains robotic models using self-supervised inverse dynamics on unlabeled data followed by lightweight language grounding, achieving superior pe…

02.07.2026 ·▲ 6

Paper LLMs & Texto

Video-Oasis: Rethinking Evaluation of Video Understanding

Video-Oasis diagnostics reveal that half of existing video benchmarks can be solved without visual input, exposing significant capability gaps in current video understanding models…

02.07.2026 ·▲ 58

Paper Visão Computacional

VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon

VLA-Corrector addresses limitations of action chunking in vision-language-action models by introducing a lightweight latent-space vision monitor that enables adaptive corrective re…

02.07.2026 ·▲ 24

Blog Multimodal

Auditing Generalization in AI-Generated Video Detection: A Six-Control Protocol and the VidAudit Toolkit

arXiv:2606.31004v1 Announce Type: new Abstract: AI-generated video detection benchmarks such as GenVidBench and AIGVDBench are the de facto leaderboards, yet most evaluation protocols leave uncontrolled confounds that can inflate reported generalization. As an existence proof, a three-feature clip-length classifier reaches a leave-one-generator-out (LOGO) AUC of 0.998 on GenVidBench under unaudited evaluation, while measuring nothing about motion. A 20-paper survey finds none applying all six st...

01.07.2026

Blog Robótica & RL

3D HAMSTER: Bridging Planning and Control in Hierarchical Vision Language Action Models through 3D Trajectory Guidance

arXiv:2606.31329v1 Announce Type: new Abstract: Hierarchical Vision-Language-Action (VLA) models decouple high-level planning from low-level control to improve generalization in robot manipulation. Recent work in this paradigm uses 2D end-effector trajectories predicted by a Vision-Language Model (VLM) as explicit guidance for a downstream policy. However, state-of-the-art low-level policies operate in 3D metric space on point clouds, and feeding them 2D guidance that lacks depth forces each way...

01.07.2026

Blog Visão Computacional

Dense Structural Priors for Sparse Functional Landmark Localization in Surgical Videos

arXiv:2606.31007v1 Announce Type: new Abstract: Vision foundation models such as SAM 3 can provide transferable object-level structure across diverse surgical video conditions, but segmentation outputs do not explicitly encode the action-conditioned semantics that define functional surgical landmarks. Estimating instrument extent and geometry differs from localizing the tip or anchor relevant to clipping, grasping, or dissecting. We investigate vision foundation model-enabled sparse action-aware...

01.07.2026