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

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

6 Jun 2024arXiv:2406.04334archive 2025-07-28

Lingchen Meng, Jianwei Yang, Rui Tian, Xiyang Dai, Zuxuan Wu, Jianfeng Gao, Yu-Gang Jiang

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.

PaperPDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Language ModellingLarge Language ModelVisual Question AnsweringZero-Shot Video Question Answer

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Visual Question Answering MM-Vet DeepStack-L-HD (Vicuna-13B) GPT-4 score 39.3 #119 of 231 Archive leaderboard report
Zero-Shot Video Question Answer NExT-QA DeepStack-L(7B) Accuracy 61.0 #21 of 27 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AttentionDense ConnectionsLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSoftmaxVision Transformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections