Leading VLMs show high cross-view consistency paired with low metric accuracy on distance queries, indicating evidence-insensitive reasoning rather than geometric grounding.
Vila: On pre-training for visual language models.arXiv preprint arXiv:2405.04464, 2024
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Determines the Ekedahl-Oort strata containing images under natural embeddings of orthogonal and unitary Shimura varieties and computes their discrete invariants.
citing papers explorer
-
Consistent Yet Wrong: Evidence Insensitivity in Spatial Vision-Language Models
Leading VLMs show high cross-view consistency paired with low metric accuracy on distance queries, indicating evidence-insensitive reasoning rather than geometric grounding.
-
Ekedahl-Oort strata under natural embeddings of orthogonal and unitary Shimura varieties
Determines the Ekedahl-Oort strata containing images under natural embeddings of orthogonal and unitary Shimura varieties and computes their discrete invariants.