DeepStack: Deeply Stacking Visual Tokens is Surprisingly Simple and Effective for LMMs

TL;DR

DeepStack introduces a simple yet effective method of stacking visual tokens across transformer layers in large multimodal models, significantly improving performance with minimal additional computational cost.

Contribution

The paper proposes DeepStack, a novel architecture that stacks visual tokens into multiple groups across layers, enhancing interaction modeling in LMMs with minimal extra cost.

Findings

DeepStack improves LMM performance by 2.7-2.9 points on 9 benchmarks.

It achieves comparable results with only one-fifth of the context length.

Significant gains on high-resolution tasks like TextVQA, DocVQA, and InfoVQA.

Abstract

Most large multimodal models (LMMs) are implemented by feeding visual tokens as a sequence into the first layer of a large language model (LLM). The resulting architecture is simple but significantly increases computation and memory costs, as it has to handle a large number of additional tokens in its input layer. This paper presents a new architecture DeepStack for LMMs. Considering $N$ layers in the language and vision transformer of LMMs, we stack the visual tokens into $N$ groups and feed each group to its aligned transformer layer \textit{from bottom to top}. Surprisingly, this simple method greatly enhances the power of LMMs to model interactions among visual tokens across layers but with minimal additional cost. We apply DeepStack to both language and vision transformer in LMMs, and validate the effectiveness of DeepStack LMMs with extensive empirical results. Using the same context length, our DeepStack 7B and 13B parameters surpass their counterparts by \textbf{2.7} and \textbf{2.9} on average across \textbf{9} benchmarks, respectively. Using only one-fifth of the context length, DeepStack rivals closely to the counterparts that use the full context length. These gains are particularly pronounced on high-resolution tasks, e.g., \textbf{4.2}, \textbf{11.0}, and \textbf{4.0} improvements on TextVQA, DocVQA, and InfoVQA compared to LLaVA-1.5-7B, respectively. We further apply DeepStack to vision transformer layers, which brings us a similar amount of improvements, \textbf{3.8} on average compared with LLaVA-1.5-7B.