ViTL uses an LLM to translate natural language into LTL for DFA-coordinated VLM navigation, enabling zero-shot long-horizon tasks with temporal constraints on HM3D.
Motion planning under temporal logic specifications in semantically unknown environments,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
ViTL: Temporal Logic-Guided Zero-Shot Natural Language Navigation via Vision-Language Models
ViTL uses an LLM to translate natural language into LTL for DFA-coordinated VLM navigation, enabling zero-shot long-horizon tasks with temporal constraints on HM3D.