ShowTable: Unlocking Creative Table Visualization with Collaborative Reflection and Refinement

TL;DR

ShowTable presents a novel pipeline combining multi-modal large language models and diffusion models for creative, data-faithful infographic generation from tables, supported by new benchmarks and automated training pipelines.

Contribution

The paper introduces ShowTable, a new pipeline that integrates reasoning and image generation models for creative table visualization, along with a benchmark and automated data construction methods.

Findings

Outperforms baseline methods in visual fidelity and accuracy

Demonstrates effective multi-modal reasoning and error correction

Provides a new benchmark with 800 challenging instances

Abstract

While existing generation and unified models excel at general image generation, they struggle with tasks requiring deep reasoning, planning, and precise data-to-visual mapping abilities beyond general scenarios. To push beyond the existing limitations, we introduce a new and challenging task: creative table visualization, requiring the model to generate an infographic that faithfully and aesthetically visualizes the data from a given table. To address this challenge, we propose ShowTable, a pipeline that synergizes MLLMs with diffusion models via a progressive self-correcting process. The MLLM acts as the central orchestrator for reasoning the visual plan and judging visual errors to provide refined instructions, the diffusion execute the commands from MLLM, achieving high-fidelity results. To support this task and our pipeline, we introduce three automated data construction pipelines for training different modules. Furthermore, we introduce TableVisBench, a new benchmark with 800 challenging instances across 5 evaluation dimensions, to assess performance on this task. Experiments demonstrate that our pipeline, instantiated with different models, significantly outperforms baselines, highlighting its effective multi-modal reasoning, generation, and error correction capabilities.