Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
Dynamic-LLaVA: Efficient Multimodal Large Language Models via Dynamic Vision-language Context Sparsification
2025-01-22
· ICLR 2025 Poster ·
anchor
Findings
IC-414
LLaVA-1.5-7B and LLaVA-1.5-13B exhibit severe performance degradation when H2O KV cache compression is applied in multimodal settings