GLANCE steers VLM agents to explore uncertain regions by rewarding the discrepancy between what their linguistic world model predicts and what they actually see.
ahead/forward
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
What You Think is What You See: Driving Exploration in VLM Agents via Visual-Linguistic Curiosity
GLANCE steers VLM agents to explore uncertain regions by rewarding the discrepancy between what their linguistic world model predicts and what they actually see.