DocSLM: A Small Vision-Language Model for Long Multimodal Document Understanding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hannan, Tanveer, Mallios, Dimitrios, Pathak, Parth, Sardari, Faegheh, Seidl, Thomas, Bertasius, Gedas, Fayyaz, Mohsen, Sengupta, Sunando
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908668850601984
author Hannan, Tanveer
Mallios, Dimitrios
Pathak, Parth
Sardari, Faegheh
Seidl, Thomas
Bertasius, Gedas
Fayyaz, Mohsen
Sengupta, Sunando
author_facet Hannan, Tanveer
Mallios, Dimitrios
Pathak, Parth
Sardari, Faegheh
Seidl, Thomas
Bertasius, Gedas
Fayyaz, Mohsen
Sengupta, Sunando
contents Large Vision-Language Models (LVLMs) have demonstrated strong multimodal reasoning capabilities on long and complex documents. However, their high memory footprint makes them impractical for deployment on resource-constrained edge devices. We present DocSLM, an efficient Small Vision-Language Model designed for long-document understanding under constrained memory resources. DocSLM incorporates a Hierarchical Multimodal Compressor that jointly encodes visual, textual, and layout information from each page into a fixed-length sequence, greatly reducing memory consumption while preserving both local and global semantics. To enable scalable processing over arbitrarily long inputs, we introduce a Streaming Abstention mechanism that operates on document segments sequentially and filters low-confidence responses using an entropy-based uncertainty calibrator. Across multiple long multimodal document benchmarks, DocSLM matches or surpasses state-of-the-art methods while using 82\% fewer visual tokens, 75\% fewer parameters, and 71\% lower latency, delivering reliable multimodal document understanding on lightweight edge devices. Code and Model are available in https://github.com/Tanveer81/DocSLM.git.
format Preprint
id arxiv_https___arxiv_org_abs_2511_11313
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DocSLM: A Small Vision-Language Model for Long Multimodal Document Understanding
Hannan, Tanveer
Mallios, Dimitrios
Pathak, Parth
Sardari, Faegheh
Seidl, Thomas
Bertasius, Gedas
Fayyaz, Mohsen
Sengupta, Sunando
Computer Vision and Pattern Recognition
Large Vision-Language Models (LVLMs) have demonstrated strong multimodal reasoning capabilities on long and complex documents. However, their high memory footprint makes them impractical for deployment on resource-constrained edge devices. We present DocSLM, an efficient Small Vision-Language Model designed for long-document understanding under constrained memory resources. DocSLM incorporates a Hierarchical Multimodal Compressor that jointly encodes visual, textual, and layout information from each page into a fixed-length sequence, greatly reducing memory consumption while preserving both local and global semantics. To enable scalable processing over arbitrarily long inputs, we introduce a Streaming Abstention mechanism that operates on document segments sequentially and filters low-confidence responses using an entropy-based uncertainty calibrator. Across multiple long multimodal document benchmarks, DocSLM matches or surpasses state-of-the-art methods while using 82\% fewer visual tokens, 75\% fewer parameters, and 71\% lower latency, delivering reliable multimodal document understanding on lightweight edge devices. Code and Model are available in https://github.com/Tanveer81/DocSLM.git.
title DocSLM: A Small Vision-Language Model for Long Multimodal Document Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.11313