Pith. sign in

Multi-modal Situated Reasoning in 3D Scenes

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

Situation awareness is essential for understanding and reasoning about 3D scenes in embodied AI agents. However, existing datasets and benchmarks for situated understanding are limited in data modality, diversity, scale, and task scope. To address these limitations, we propose Multi-modal Situated Question Answering (MSQA), a large-scale multi-modal situated reasoning dataset, scalably collected leveraging 3D scene graphs and vision-language models (VLMs) across a diverse range of real-world 3D scenes. MSQA includes 251K situated question-answering pairs across 9 distinct question categories, covering complex scenarios within 3D scenes. We introduce a novel interleaved multi-modal input setting in our benchmark to provide text, image, and point cloud for situation and question description, resolving ambiguity in previous single-modality convention (e.g., text). Additionally, we devise the Multi-modal Situated Next-step Navigation (MSNN) benchmark to evaluate models' situated reasoning for navigation. Comprehensive evaluations on MSQA and MSNN highlight the limitations of existing vision-language models and underscore the importance of handling multi-modal interleaved inputs and situation modeling. Experiments on data scaling and cross-domain transfer further demonstrate the efficacy of leveraging MSQA as a pre-training dataset for developing more powerful situated reasoning models.

fields

cs.CV 1

years

2025 1

verdicts

CONDITIONAL 1

representative citing papers

Hypo3D: Exploring Hypothetical Reasoning in 3D

cs.CV · 2025-02-02 · conditional · novelty 7.0

Hypo3D, a 3D VQA benchmark with 14,885 QAs over 700 indoor scenes, tests whether models can reason about imagined scene changes and shows they fall far behind humans.

citing papers explorer

Showing 1 of 1 citing paper.

  • Hypo3D: Exploring Hypothetical Reasoning in 3D cs.CV · 2025-02-02 · conditional · none · ref 25 · internal anchor

    Hypo3D, a 3D VQA benchmark with 14,885 QAs over 700 indoor scenes, tests whether models can reason about imagined scene changes and shows they fall far behind humans.