STAC compresses KV caches in streaming 3D reconstruction transformers via temporal token preservation with decayed attention, spatial voxel compression, and chunked multi-frame optimization, delivering 10x memory reduction and 4x faster inference at SOTA quality.
Vi- sion transformers for dense prediction
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2representative citing papers
Leading VLMs show high cross-view consistency paired with low metric accuracy on distance queries, indicating evidence-insensitive reasoning rather than geometric grounding.
citing papers explorer
-
STAC: Plug-and-Play Spatio-Temporal Aware Cache Compression for Streaming 3D Reconstruction
STAC compresses KV caches in streaming 3D reconstruction transformers via temporal token preservation with decayed attention, spatial voxel compression, and chunked multi-frame optimization, delivering 10x memory reduction and 4x faster inference at SOTA quality.
-
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.