Search papers, labs, and topics across Lattice.
This paper introduces FreqCache, a frequency-guided token caching framework designed to accelerate Vision-Language-Navigation (VLN) models by addressing the limitations of existing visual domain token caching methods. FreqCache leverages the frequency domain to optimize token cache establishment, refreshment, and adaptive adjustment, overcoming challenges related to viewpoint migration, edge information neglect, and temporal variation. Experiments demonstrate that FreqCache achieves a 1.59x speedup with minimal overhead, highlighting the effectiveness of frequency domain methods in VLN token caching.
Frequency domain analysis unlocks 1.59x speedups in Vision-Language-Navigation by enabling optimal token caching, a feat previously limited by visual domain approaches.
Vision-Language-Navigation (VLN) models exhibit excellent navigation accuracy but incur high computational overhead. Token caching has emerged as a promising training-free strategy to reduce this cost by reusing token computation results; however, existing token caching approaches rely on visual domain methods for cacheable token selection, leading to challenges when adapted to VLN models. 1) Visual domain methods become invalid when there is viewpoint migration. 2) Visual domain methods neglect critical edge information without the aid of additional algorithms. 3) Visual domain methods overlook the temporal variation of scenarios and lack adjustability in cache budgets. In this paper, we develop detailed analyses and find that the impacts of these challenges exhibit invariance and analyzability in the frequency domain. Based on these, we propose a frequency-guided token caching framework, called FreqCache. Utilizing the inherent properties of the frequency domain, FreqCache achieves optimal token cache establishment, refreshment, and adaptive adjustment. Experiments show that FreqCache achieves 1.59x speedup with ignorable overhead, showing the effect of integrating frequency domain methods in VLN token caching.
