@inproceedings{ravi-etal-2025-sight,
    title = "Out of Sight, Not Out of Context? Egocentric Spatial Reasoning in {VLM}s Across Disjoint Frames",
    author = "Ravi, Sahithya  and
      Sarch, Gabriel Herbert  and
      Vineet, Vibhav  and
      Wilson, Andrew D  and
      Kumaravel, Balasaravanan Thoravi",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.816/",
    pages = "16146--16161",
    ISBN = "979-8-89176-332-6",
    abstract = "An embodied AI assistant operating on egocentric video must integrate spatial cues across time - for instance, determining where an object A, glimpsed a few moments ago lies relative to an object B encountered later. We introduce Disjoint-3DQA , a generative QA benchmark that evaluates this ability of VLMs by posing questions about object pairs that are not co-visible in the same frame. We evaluated seven state-of-the-art VLMs and found that models lag behind human performance by 28{\%}, with steeper declines in accuracy (60{\%} {\textrightarrow} 30 {\%}) as the temporal gap widens. Our analysis further reveals that providing trajectories or bird{'}s-eye-view projections to VLMs results in only marginal improvements, whereas providing oracle 3D coordinates leads to a substantial 20{\%} performance increase. This highlights a core bottleneck of multi-frame VLMs in constructing and maintaining 3D scene representations over time from visual signals. Disjoint-3DQA therefore sets a clear, measurable challenge for long-horizon spatial reasoning and aims to catalyze future research at the intersection of vision, language, and embodied AI."
}