SpaMEM benchmark shows multimodal LLMs succeed at spatial tasks with text histories but sharply fail at long-horizon belief maintenance from raw visual streams alone.
arXiv preprint arXiv:2503.11089 (2025),https: //arxiv.org/abs/2503.110894
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
SpaMEM: Benchmarking Dynamic Spatial Reasoning via Perception-Memory Integration in Embodied Environments
SpaMEM benchmark shows multimodal LLMs succeed at spatial tasks with text histories but sharply fail at long-horizon belief maintenance from raw visual streams alone.