TableDART: Dynamic Adaptive Multi-Modal Routing for Table Understanding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xing, Xiaobo, Yuan, Wei, Chen, Tong, Nguyen, Quoc Viet Hung, Zhang, Xiangliang, Yin, Hongzhi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915789665206272
author Xing, Xiaobo
Yuan, Wei
Chen, Tong
Nguyen, Quoc Viet Hung
Zhang, Xiangliang
Yin, Hongzhi
author_facet Xing, Xiaobo
Yuan, Wei
Chen, Tong
Nguyen, Quoc Viet Hung
Zhang, Xiangliang
Yin, Hongzhi
contents Modeling semantic and structural information from tabular data remains a core challenge for effective table understanding. Existing Table-as-Text approaches flatten tables for large language models (LLMs), but lose crucial structural cues, while Table-as-Image methods preserve structure yet struggle with precise semantics. Recent Table-as-Multimodality strategies attempt to combine textual and visual views, but they (1) statically process both modalities for every query-table pair within large multimodal LLMs (MLLMs), inevitably introducing redundancy and even conflicts, and (2) depend on costly fine-tuning of MLLMs. In light of this, we propose TableDART, a training-efficient framework that integrates multimodal views by reusing pretrained single-modality models. TableDART introduces a lightweight 2.59M-parameter MLP gating network that dynamically selects the optimal path (Text-only, Image-only, or Fusion) for each table-query pair, reducing redundancy and avoiding conflicts that arise when textual and visual views of the same table provide inconsistent cues. By routing to the most appropriate view, our framework improves both accuracy and efficiency. In addition, we propose a novel agent to mediate cross-modal knowledge integration by analyzing outputs from text- and image-based models, either selecting the best result or synthesizing a new answer through reasoning. This design avoids the prohibitive costs of full MLLM fine-tuning. Extensive experiments on seven benchmarks show that TableDART establishes new state-of-the-art performance among open-source models, surpassing the strongest baseline by an average of 4.02%. The code is available at: https://github.com/xiaobo-xing/TableDART.
format Preprint
id arxiv_https___arxiv_org_abs_2509_14671
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TableDART: Dynamic Adaptive Multi-Modal Routing for Table Understanding
Xing, Xiaobo
Yuan, Wei
Chen, Tong
Nguyen, Quoc Viet Hung
Zhang, Xiangliang
Yin, Hongzhi
Computation and Language
Artificial Intelligence
Machine Learning
Modeling semantic and structural information from tabular data remains a core challenge for effective table understanding. Existing Table-as-Text approaches flatten tables for large language models (LLMs), but lose crucial structural cues, while Table-as-Image methods preserve structure yet struggle with precise semantics. Recent Table-as-Multimodality strategies attempt to combine textual and visual views, but they (1) statically process both modalities for every query-table pair within large multimodal LLMs (MLLMs), inevitably introducing redundancy and even conflicts, and (2) depend on costly fine-tuning of MLLMs. In light of this, we propose TableDART, a training-efficient framework that integrates multimodal views by reusing pretrained single-modality models. TableDART introduces a lightweight 2.59M-parameter MLP gating network that dynamically selects the optimal path (Text-only, Image-only, or Fusion) for each table-query pair, reducing redundancy and avoiding conflicts that arise when textual and visual views of the same table provide inconsistent cues. By routing to the most appropriate view, our framework improves both accuracy and efficiency. In addition, we propose a novel agent to mediate cross-modal knowledge integration by analyzing outputs from text- and image-based models, either selecting the best result or synthesizing a new answer through reasoning. This design avoids the prohibitive costs of full MLLM fine-tuning. Extensive experiments on seven benchmarks show that TableDART establishes new state-of-the-art performance among open-source models, surpassing the strongest baseline by an average of 4.02%. The code is available at: https://github.com/xiaobo-xing/TableDART.
title TableDART: Dynamic Adaptive Multi-Modal Routing for Table Understanding
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2509.14671