JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in Robotics
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (7) · 5276-5286
DOI: 10.1609/aaai.v40i7.37443 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Simindokht Jahangard und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Recent advances in Vision-Language Models (VLMs) and large language models (LLMs) have greatly enhanced visual reasoning, a key capability for embodied AI agents like robots. However, existing visual reasoning benchmarks often suffer from several limitations: they lack a clear definition of reasoning complexity, offer have no control to generate questions over varying difficulty and task customization, and fail to provide structured, step-by-step reasoning annotations (workflows). To bridge these gaps, we formalize reasoning complexity, introduce an adaptive query engine that generates customizable questions of varying complexity with detailed intermediate annotations, and extend the JRDB dataset with human-object interaction and geometric relationship annotations to create JRDB-Reasoning, a benchmark tailored for visual reasoning in human-crowded environments. Our engine and benchmark enable fine-grained evaluation of visual reasoning frameworks and dynamic assessment of visual-language models across reasoning levels.
Zitieren
APA
Jahangard, S., Mohammadi, M., Shen, Y., Cai, Z., & Rezatofighi, H. (2026). JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in Robotics. Proceedings of the AAAI Conference on Artificial Intelligence, 40(7), 5276-5286. https://doi.org/10.1609/aaai.v40i7.37443
Vancouver
Jahangard S, Mohammadi M, Shen Y, Cai Z, Rezatofighi H. JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in Robotics. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(7):5276-5286. doi:10.1609/aaai.v40i7.37443
BibTeX
@article{Jahangard2026,
author = {Jahangard, Simindokht and Mohammadi, Mehrzad and Shen, Yi and Cai, Zhixi and Rezatofighi, Hamid},
title = {JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in Robotics},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {7},
pages = {5276--5286},
doi = {10.1609/aaai.v40i7.37443}
}
EUVIMED