OMEGA: Optimized Multimodal Position Encoding Index Derivation with Global Adaptive Scaling for Vision-Language Models
This addresses a bottleneck in VLMs for multimodal AI applications, offering incremental improvements over existing methods.
The paper tackled the problem of suboptimal position encoding in Vision-Language Models (VLMs) by proposing OMEGA, a framework with modality-specific encoding and adaptive scaling, which improved VLM performance by up to 3.43% on visual-intensive tasks.
Vision-Language Models (VLMs) have demonstrated strong performance across various multimodal tasks, where position encoding plays a vital role in modeling both the sequential structure of textual information and the spatial structure of visual information. However, current VLMs commonly adopt modality-unified 1D or 2D positional indexing strategies, which treat textual and visual tokens uniformly without accounting for their distinct structural properties and sequential continuity for text and spatial coherence for vision. To address this limitation, we propose OMEGA, a novel position encoding framework that employs Modality-Specific Position Encoding (MSPE) to assign positional indices while preserving the inherent structures of each modality across separate coordinate dimensions. Additionally, to align the information density of multimodal data in the positional index space, OMEGA introduces Global Adaptive Encoding Step Scaling (GAESS), which adaptively adjusts the position encoding step size of visual tokens based on the embedding entropy of both modalities. Experimental results demonstrate that OMEGA consistently enhances VLM performance across diverse architectures and VQA benchmarks. On visual-intensive tasks, OMEGA achieves up to 3.43% improvement over baseline position encoding strategies on Qwen2.5-VL-3B, with consistent gains observed across larger models including Qwen2.5-VL-7B and LLaVA-v1.5-7B.