Scene-Aware Dialogue
4 papers with code • 1 benchmarks • 1 datasets
AVSD
Most implemented papers
A Simple Baseline for Audio-Visual Scene-Aware Dialog
idansc/simple-avsd
•
•
CVPR 2019
The recently proposed audio-visual scene-aware dialog task paves the way to a more data-driven way of learning virtual assistants, smart speakers and car navigation systems.
Maintaining Common Ground in Dynamic Environments
Alab-NII/dynamic-onecommon
•
•
Common grounding is the process of creating and maintaining mutual understandings, which is a critical aspect of sophisticated human communication.
An Embodied Generalist Agent in 3D World
embodied-generalist/embodied-generalist
•
•
However, several significant challenges remain: (i) most of these models rely on 2D images yet exhibit a limited capacity for 3D input; (ii) these models rarely explore the tasks inherently defined in 3D world, e. g., 3D grounding, embodied reasoning and acting.
