chart-understanding

mPLUG-DocOwl: Modularized Multimodal Large Language Model for Document Understanding

A curated list of recent and past chart understanding work based on our IEEE TKDE survey paper: From Pixels to Insights: A Survey on Automatic Chart Understanding in the Era of Large Foundation Models.

[NeurIPS 2024] CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs

[ICLR2025 Oral] ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding

Official Implementation of "Visual-ERM: Reward Modeling for Visual Equivalence"

Code and data for the ACL 2024 Findings paper "Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning"

[NeurIPS 2025] ChartMuseum: Testing Visual Reasoning Capabilities of Large Vision-Language Models

A curated collection of papers, datasets, and resources on the topic of misleading chart understanding

ChartHal: A Fine-grained Framework Evaluating Hallucination of Large Vision Language Models in Scientific Chart Understanding

Code associated with the ACL 2026 Main conference paper: "Is this chart lying to me? Automating the detection of misleading visualization"

SciVQA: Scientific Visual Question Answering shared task

This is the official repository for our paper 📄 “In-Depth and In-Breadth: Pre-training Multimodal Language Models Customized for Comprehensive Chart Understanding”

VisDoTQA is a chart visual reasoning benchmark and dataset for Position, Length, Pattern, and Extract tasks.

Chart VQA is an intelligent chart question-answering system designed to help users understand information presented in charts and graphs. The project combines computer vision, optical character recognition, and vision-language models to analyze an uploaded chart and answer questions about its content in Vietnamese or English.

CVLAT: a diagnostic benchmark for visual–factual arbitration in LVLMs' visualization literacy

Chart-to-table extraction results on the ChartX benchmark — extracted tables released for independent verification.

Chart as Knowledge Encoding: A Unified Framework for AI Visual Intelligence