OpenEQA: Embodied Question Answering in the Era of Foundation Models

By Arjun Majumdar, Anurag Ajay, Xiaohan Zhang, Pranav Putta, Sriram Yenamandra, Mikael Henaff, S. Silwal, Paul Mcvay, Oleksandr Maksymets, Sergio Arnaud and 14 more. Computer Vision and Pattern Recognition, 2024

348 citations recorded by the source corpus.

Subfields: Artificial Intelligence

View Paper: https://www.semanticscholar.org/paper/920731b589af90a5b79236f4939ac117bbb939f2

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yann LeCun
2. Yoshua Bengio
3. Saining Xie
4. Quoc V. Le
5. Trevor Darrell
6. Jiashi Feng
7. Jan Kautz
8. Rob Fergus

Citations in the corpus

• Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces (2024, positive): “With MLLMs displaying impressive performance on still-images across perception, reasoning, and multi-disciplinary tasks [39, 52, 98, 99], there is increasing interest in evaluating MLLMs’ video understanding capabilities [23, 24, 43, 44, 47, 51, 54, 55, 63, 85, 96].”
• SpatialRGPT: Grounded Spatial Reasoning in Vision Language Model (2024, methodology): “Yet, recent studies [9] indicate that LLMs struggle to utilize coordinate information effectively when presented in text, which can undermine their ability to understand and reason about spatial relationships.”
• Aligning Cyber Space With Physical World: A Comprehensive Survey on Embodied AI (2024, positive): “The key to this approach lies in abstract processing and knowledge decoupling, enabling efficient adaptation to complex situations.”
• Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs (2024, background): “Determining whether a benchmark truly needs visual input to be solved has been a persistent challenge in vision-language research [2, 21, 43, 87].”
• A Survey on Vision--Language--Action Models for Embodied AI (2024, positive): “OpenEQA [230] evaluates seven main categories, including functional reasoning and world knowledge.”
• eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents (2026, unknown): “Embodied-QA benchmarks such as OpenEQA [Majumdar et al., 2024] test perception-grounded question answering in scenes, but the questions are surface-task defined (“what colour is the chair?”) and do not isolate which memory function produces a correct or incorrect answer.”