10 papers
StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modeling
Meng Wei, Chenyang Wan, Xiqian Yu +9
Vision-and-Language Navigation (VLN) in real-world settings requires agents to process continuous visual streams and generate actions with low latency grounded in language instruct…
GTAM: Geometry Grounded Track Anything Model
Chenming Zhu, Peizhou Cao, Jingli Lin +5
Human spatial understanding arises from jointly perceiving geometry and semantics, enabling consistent object identification and localization across viewpoints and time. Current vi…
Thinking with Imagination: Agentic Visual Spatial Reasoning with World Simulators
Chenming Zhu, Jingli Lin, Yilin Long +4
While Vision-Language Models (VLMs) have shown strong visual reasoning capabilities, their spatial reasoning abilities remain largely constrained to the observed images and text-or…
MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence
Sihan Yang, Runsen Xu, Yiman Xie +10
Spatial intelligence is essential for multimodal large language models (MLLMs) operating in the complex physical world. Existing benchmarks, however, probe only single-image relati…
MMSI-Video-Bench: A Holistic Benchmark for Video-Based Spatial Intelligence
Jingli Lin, Runsen Xu, Shaohao Zhu +11
Spatial understanding over continuous visual input is crucial for MLLMs to evolve into general-purpose assistants in physical environments. Yet there is still no comprehensive benc…
Ground Slow, Move Fast: A Dual-System Foundation Model for Generalizable Vision-and-Language Navigation
Meng Wei, Chenyang Wan, Jiaqi Peng +8
While recent large vision-language models (VLMs) have improved generalization in vision-language navigation (VLN), existing methods typically rely on end-to-end pipelines that map…