PARCEL is a new visual tokenization architecture combining pool-anchored resampling with conditioned elastic queries to enhance performance-efficiency tradeoffs in LVLMs over prior matryoshka methods.
Energy-driven adaptive visual token pruning for efficient vision-language models
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
FreqCache uses frequency domain properties to adaptively select, refresh, and budget token caches in VLN models, delivering 1.59x speedup with negligible overhead.
A roadmap that defines architectural nativity for multimodal models and categorizes them into Multi-to-Text, Multi-to-Target, and Multi-to-Multi types while outlining an industrial pipeline toward unified transformer-based native multimodal modeling.
citing papers explorer
-
FreqCache: Accelerating Embodied VLN Models with Adaptive Frequency-Guided Token Caching
FreqCache uses frequency domain properties to adaptively select, refresh, and budget token caches in VLN models, delivering 1.59x speedup with negligible overhead.