Embodied3DBench: Benchmarking Low-Level Embodied Spatial Intelligence of Vision Language Models
Open Access
- 9 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 14488-14497
Abstract
No abstract availableFunding Information
- Beijing Natural Science Foundation (L2608141)
- National Natural Science Foundation of China (62136001)
This publication has 20 references indexed in Scilit:
- RoboBrain: A Unified Brain Model for Robotic Manipulation from Abstract to Concrete2025
- Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces2025
- GEAL: Generalizable 3D Affordance Learning with Cross-Modal Consistency2025
- OmniManip: Towards General Robotic Manipulation via Object-Centric Interaction Primitives as Spatial Constraints2025
- RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics2025
- SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities2024
- GAPartNet: Cross-Category Domain-Generalizable Object Perception and Manipulation via Generalizable and Actionable Parts2023
- Domain Randomization-Enhanced Depth Simulation and Restoration for Perceiving and Grasping Specular and Transparent Objects2022
- PartNet: A Large-Scale Benchmark for Fine-Grained and Hierarchical Part-Level 3D Object Understanding2019
- 3D Bounding Box Estimation Using Deep Learning and Geometry2017