https://ym-xu.github.io/living-survey/fields/document-understanding/#tag=chartsUnderstanding and generating charts, plots, scientific figures and infographics.Thu, 12 Nov 2026 15:12:01 +0000https://arxiv.org/abs/2611.03023https://arxiv.org/abs/2611.03023Mon, 02 Nov 2026 00:00:00 +0000Haiyang Zhang, Sicheng Song, Zixin Chen, Jie Gu, et al. Misleading charts can lead vision-language models to conclusions that conflict with underlying data. Existing work focuses on inference-time strategies, while training-time supervision remains less explored. We present VizDAGuide, a DAG-structured supervision method for misleading-chart question answering. The method r…https://arxiv.org/abs/2611.00092https://arxiv.org/abs/2611.00092Thu, 22 Oct 2026 00:00:00 +0000Aman Rangapur, Antonio Jimeno Yepes, Daniel Bustamante, Nutchapol Dendumrongsup, et al. We introduce OCRia-1, a 4B vision-language model (VLM) for document parsing that excels at preserving table structure, extracting chart data and retaining text formatting. OCRia-1 handles content that has traditionally been challenging in document parsing: merged cells and spanning headers, tables with thousands of cel…https://arxiv.org/abs/2610.28282https://arxiv.org/abs/2610.28282Tue, 20 Oct 2026 00:00:00 +0000Qi Wang, Peijie Wang, Jun Cao, Haoyang Tong, et al. Multimodal large language models (MLLMs) have substantially advanced chart understanding and reasoning, yet their performance remains highly uneven across chart types. While modern MLLMs perform well on common charts, their accuracy drops sharply on long-tail charts with diverse visual structures and reasoning requirem…https://arxiv.org/abs/2610.18763https://arxiv.org/abs/2610.18763Tue, 13 Oct 2026 00:00:00 +0000Keshav Gupta, Jeremy Owen, Christopher Lloyd, Shaunak Mishra, et al. Online learning platforms are increasingly deploying AI assistants to provide personalized, on-demand tutoring. These can leverage infographics to communicate complex information and improve the learning experience. However, automating effective infographic generation remains a challenging task, requiring a combination…https://arxiv.org/abs/2610.13502https://arxiv.org/abs/2610.13502Fri, 09 Oct 2026 00:00:00 +0000Yan Zhou, Yang Yang Infographic generation challenges text-to-image models because a single image must satisfy numerous visual, textual, and semantic constraints. Iterative agentic refinement with a multimodal verifier is a natural remedy, but existing methods face a scale mismatch: local edits cannot correct global composition errors, wh…https://arxiv.org/abs/2610.08106https://arxiv.org/abs/2610.08106Tue, 06 Oct 2026 00:00:00 +0000Langxi Huang, Pingping Zhang, Lanyun Zhu, Chunyang Jiang, et al. Multimodal large language models (MLLMs) advance rapidly, while conventional benchmark development lags behind, delaying investigation of newly observed capability gaps. Such investigation requires an expressive task format and an on-demand construction process: information-rich charts make chart question answering (Ch…https://arxiv.org/abs/2610.00374https://arxiv.org/abs/2610.00374Wed, 30 Sep 2026 00:00:00 +0000Yuxin Yue, Yingchen Zhang, Ruqing Zhang, Jiafeng Guo, et al. Analytical charts in multimodal deep research encode quantitative claims, requiring every visualized value to be faithfully grounded in supporting evidence. Unlike retrieved images that mainly provide contextual information, charts require numerical fidelity: visualized values should not only match retrieved evidence q…https://arxiv.org/abs/2609.39380https://arxiv.org/abs/2609.39380Wed, 30 Sep 2026 00:00:00 +0000Yifan Li, Tong Li, Qi Zeng, Lishuai Gao, et al. Reliable infographic generation requires facts, symbols, and visual relations to remain consistent through rendering and revision. Correcting one element also requires tracking its supporting evidence and the dependencies affected by the change. We present \textbf{InfoAgent}, a training-free framework for \emph{evidenc…https://arxiv.org/abs/2609.38781https://arxiv.org/abs/2609.38781Wed, 30 Sep 2026 00:00:00 +0000Xinhe Wu, Yadong Jin Multimodal large language models (MLLMs) offer a promising approach for recovering numerical data from scientific charts, but their ability to reconstruct chart data from visually dense figures remains poorly understood. Existing chart understanding benchmarks primarily evaluate question answering or chart-level reason…https://arxiv.org/abs/2609.38642https://arxiv.org/abs/2609.38642Tue, 29 Sep 2026 00:00:00 +0000Jiaxiang Tang, Yi Zhou, Chad DeLuca, Rogerio Feris, et al. Chart editing requires cross-modal edit grounding, realizing a requested visual change in the code that draws it, with necessary related updates and without altering unrelated content. Existing benchmarks emphasize either code executability or chart quality, but their metrics do not clearly distinguish request completi…https://arxiv.org/abs/2609.33286https://arxiv.org/abs/2609.33286Sun, 27 Sep 2026 00:00:00 +0000Cheng Yang, Chufan Shi, Huijuan Wang, Bo Shui, et al. Multimodal foundation models edit natural photographs at production quality, yet the same models struggle with structured visual content such as infographics. Unlike photographs, infographics encode information through logical relations; editing one element often requires surrounding elements to be adapted. We refer to…https://arxiv.org/abs/2609.24210https://arxiv.org/abs/2609.24210Mon, 21 Sep 2026 00:00:00 +0000Lijian Wu, Henry Hengyuan Zhao, Zijian Zhang, Jiahao Tang, et al. Building strong chart-to-code systems increasingly relies on reinforcement learning, whose effectiveness depends critically on the quality of the reward signal. Large Multimodal Models (LMMs) play a natural critical role in jointly assessing chart visual appearance and task requirements. They are therefore increasingly…https://arxiv.org/abs/2609.24172https://arxiv.org/abs/2609.24172Mon, 21 Sep 2026 00:00:00 +0000Xinnuo Zhang, Zhike Tang, Jing Xu, Haoyuan Zhao, et al. Legends are fundamental to chart understanding, as reliable interpretation requires correctly binding legend entries to corresponding visual marks. While vision-language models (VLMs) are increasingly applied to chart understanding, their legend understanding is poorly diagnosed by aggregate accuracy, which can be sati…https://arxiv.org/abs/2609.24071https://arxiv.org/abs/2609.24071Mon, 21 Sep 2026 00:00:00 +0000Sanchit Sinha, Oana Frunza, Kashif Rasul, Aidong Zhang · EMNLP 2026 Findings Chart reasoning agents are increasingly used to extract actionable insights in critical domains, achieving state-of-the-art performance on multiple benchmarks. Yet, high benchmark accuracy alone is insufficient for deployment, where stakeholders must be able to audit and verify how a model reaches its answer. Existing …https://arxiv.org/abs/2609.18844https://arxiv.org/abs/2609.18844Wed, 16 Sep 2026 00:00:00 +0000Liyang Fan, Chi Wei, Yitai Li, Xinping Bi, et al. Multimodal coding agents are expected to turn visual inputs into usable artifacts, and they act through a harness, the layer of tools, context management, and execution environment around the model. Existing evaluations often isolate short tool calls, API traces, or screenshot resemblance, and a low score under these p…https://arxiv.org/abs/2609.14302https://arxiv.org/abs/2609.14302Sun, 13 Sep 2026 00:00:00 +0000Xiaoya Wang, Yutong Xu, Junjie Wang · EMNLP Findings Can financial vision-language models (VLMs) turn chart evidence into reliable action recommendations? Existing hallucination evaluations are mostly claim-centric; they assess whether generated statements are supported, but not whether evidence remains traceable through rationale, confidence, and final action. We introd…https://arxiv.org/abs/2609.08657https://arxiv.org/abs/2609.08657Tue, 08 Sep 2026 00:00:00 +0000Xiaochuan Zhong, Yifan Hou, Chenxi Pang, Shaobo Cui Charts are structured visual compositions whose elements have distinct functional roles, semantic correspondences, and visibility relations. This structural view motivates evaluating whether models can understand and manipulate charts at the layer level. Existing chart benchmarks, however, primarily assess the correctn…https://arxiv.org/abs/2609.13268https://arxiv.org/abs/2609.13268Mon, 07 Sep 2026 00:00:00 +0000Amirul Rahman, Aisha Karim, Kenji Nakamura, Yi-Fan Ng Annual reports, diligence packs, and infographic dashboards bury numbers in page images: axes, cell grids, and footnotes that OCR pipelines flatten and that page-level visual retrievers still treat as interchangeable in-context examples. We keep a frozen Qwen2.5-VL-7B-Instruct generator and a ColPali / VisRAG-Ret page …https://arxiv.org/abs/2609.13267https://arxiv.org/abs/2609.13267Mon, 07 Sep 2026 00:00:00 +0000Alberlucia Rafael Soarez, Camila Ferreira, Daniel Kim, Mariana Costa, et al. Scientific charts encode quantities in axes, legends, and geometric marks, yet large vision-language models still treat them as natural photographs. Visual in-context examples do not expose the coordinate frame; unconstrained chain-of-thought can name a plausible number that was never read from a bar. We present STEER …https://arxiv.org/abs/2608.26757https://arxiv.org/abs/2608.26757Thu, 27 Aug 2026 00:00:00 +0000Jiahui tang, Kuicai Dong, Dexun Li, Hongchao Gu, et al. Faithful chart generation in real-world data-science workflows requires grounding visualizations in scattered evidence, computing chart-ready quantities, and rendering them accurately. Modern LLMs can produce visually plausible, instruction-compliant charts, yet data-level hallucinations remain difficult to detect in l…https://arxiv.org/abs/2608.26544https://arxiv.org/abs/2608.26544Thu, 27 Aug 2026 00:00:00 +0000Jinning Cui, Lu Chen, Haoyan Shi, Yue He, et al. We present Chart2SVG, a multimodal large language model that converts static raster charts into structurally organized, semantically enriched SVGs that support programmatic editing. By incorporating chart-specific semantic tokens into a vision-language model, Chart2SVG captures both geometric primitives and their funct…https://arxiv.org/abs/2608.17414https://arxiv.org/abs/2608.17414Tue, 18 Aug 2026 00:00:00 +0000Yuanbang Liu, Chenxi Ruan, Yihan Hou, Qiong Luo, et al. Chart editing requires inferring and modifying visualization code from a reference chart image based on an editing instruction, challenging fine-grained visual reasoning, instruction following, and executable code synthesis capabilities of MLLMs. Large reasoning models (LRMs) with extended Chain-of-Thought (CoT) reason…https://arxiv.org/abs/2608.17205https://arxiv.org/abs/2608.17205Mon, 17 Aug 2026 00:00:00 +0000Rodela Ghosh, Aviral Gupta, Guangjing Wang Vision-language models (VLMs) combine images and text, but when the two conflict and one becomes harder to read, it is unclear how a model shifts its reliance between them. We study this modality reallocation with a controlled setup: we degrade either the image or the text across four levels of legibility while keeping…https://arxiv.org/abs/2608.13766https://arxiv.org/abs/2608.13766Thu, 13 Aug 2026 00:00:00 +0000Mahsa Khoshnoodi, Sarah Adel Bargal Vision-language models (VLMs) remain unreliable on chart questions that require reasoning over visual quantities, and this weakness is usually attributed to a reasoning deficit and addressed with more reasoning supervision. We ask whether the difficulty lies in reasoning itself, or in the simpler skills that reasoning …https://arxiv.org/abs/2608.10677https://arxiv.org/abs/2608.10677Tue, 11 Aug 2026 00:00:00 +0000Suhaas Garre, Chris Mutty, Sushant Mehta, Edwin Chen · ECCV 2026 Workshop Professionals across medicine, engineering, finance, manufacturing, and the sciences often make consequential decisions from charts. Existing chart benchmarks do not sufficiently measure this ability: they are dominated by bar, line, and pie formats, rely on shorter reasoning chains, and are nearing saturation, with fr…https://arxiv.org/abs/2608.03464https://arxiv.org/abs/2608.03464Tue, 04 Aug 2026 00:00:00 +0000Zhenghan Chen, Zekai Shao, Lidan Tan, Xin Lin, et al. Annotations are essential to communicative visualization, helping explain data, emphasize key findings, and guide attention. While multimodal large language models (MLLMs) offer new opportunities for automatic chart annotation authoring, their capabilities in this task remain underexplored. To address this gap, we intr…https://arxiv.org/abs/2608.02833https://arxiv.org/abs/2608.02833Mon, 03 Aug 2026 00:00:00 +0000Xuehang Guo, Pingyue Zhang, Ruiyi Zhang, Zhenhailong Wang, et al. Chart question answering (CQA) requires multimodal large language models (MLLMs) to integrate visual comprehension with logical reasoning, yet current models struggle with accurate visual grounding and coherent reasoning chains. While extrinsic chain-of-thought prompting and visual cues significantly improve performanc…https://arxiv.org/abs/2608.01328https://arxiv.org/abs/2608.01328Sun, 02 Aug 2026 00:00:00 +0000Ziyan Xiao, Yinghao Zhu, Wenting Zhang, Heaju Kim, et al. Multimodal large language models (MLLMs) are rapidly evolving with expanded context windows and stronger reasoning capabilities, enabling multi-chart understanding and multi-step inference. These abilities are increasingly important as MLLMs are adopted in complex agentic tasks. However, existing benchmarks largely emp…https://arxiv.org/abs/2607.27084https://arxiv.org/abs/2607.27084Wed, 29 Jul 2026 00:00:00 +0000Zihan Deng, Chuanzhi Xu, Huiqi Liang, Haoyang Li, et al. Scientific figure quality is bound to the manuscript: a crop can be visually clear and still contradict its caption or the paragraph that cites it. Most image quality assessment and chart-understanding methods target a detached visual surface or a question-answering score, rather than alignment among the figure, its ca…https://arxiv.org/abs/2607.16060https://arxiv.org/abs/2607.16060Fri, 17 Jul 2026 00:00:00 +0000Meijia Huang, Yingjie Yin, Shihao Wang, Chenguang Ma Artistic charts combine data visualization with expressive marks, textures, and typography, but they are difficult for image generators: an output is useful only when its stylization preserves chart geometry, exact in-image text, and the semantic binding between labels and marks. We introduce ArtChart, a framework for …https://arxiv.org/abs/2607.12857https://arxiv.org/abs/2607.12857Tue, 14 Jul 2026 00:00:00 +0000Jhen-Ke Lin A generated rhythm-game chart need not reproduce one official note sequence: many note choices can fit the same song and difficulty. Reference-note agreement therefore measures reconstruction, not the full design problem. We introduce ChartGenEval, a six-question evaluation framework with an automatic, corruption-teste…https://arxiv.org/abs/2607.11192https://arxiv.org/abs/2607.11192Mon, 13 Jul 2026 00:00:00 +0000Suhaas Garre, Emily Ritchie, Sushant Mehta, Edwin Chen · CVPR 2026 Workshop A large share of day-to-day work in professional domains happens inside PDF files: benefits packets, leases, datasheets, clinical guidelines, construction plans. Benchmarks for document AI have generally measured the required capabilities in isolation: OCR, layout analysis, chart reasoning, table QA, document VQA. A hi…https://arxiv.org/abs/2607.10400https://arxiv.org/abs/2607.10400Sat, 11 Jul 2026 00:00:00 +0000Abhigya Verma, Khyati Mahajan, Amit Kumar Saha, Shruthan Radhakrishna, et al. Vision language models (VLMs) have achieved strong performance on visual document understanding benchmarks such as DocVQA, ChartQA, and MMLongBench-Doc. However, real-world documents combine multiple factors such as length, layout complexity, modality, and question difficulty, which makes it difficult to attribute mode…https://arxiv.org/abs/2607.10301https://arxiv.org/abs/2607.10301Sat, 11 Jul 2026 00:00:00 +0000Jiakang Yu, Yixuan Chai, Tianci Wang, Rihui Jin, et al. Generative image editing models struggle with structured statistical charts when data modifications require geometric synchronization. We formalize this task as Visuo-Logical Cascading Editing (VLCE). However, existing methods remain confined to localized text substitutions and struggle with dependency-aware cascading …https://arxiv.org/abs/2607.07179https://arxiv.org/abs/2607.07179Wed, 08 Jul 2026 00:00:00 +0000Miguel Lopez-Duran, Elena Marrero, Julian Fierrez, Marta Robledo-Moreno, et al. · ICDAR 2026 Workshop Document Visual Question Answering (DocVQA) presents a complex multimodal challenge, requiring models to exploit visual, textual, and layout information from documents. Although Vision-Language Models (VLMs) have shown remarkable performance in text-vision tasks, their robustness and transferability to different docume…https://arxiv.org/abs/2606.29808https://arxiv.org/abs/2606.29808Mon, 29 Jun 2026 00:00:00 +0000Yuchen He, Peizhi Ying, Liqi Cheng, Kuilin Peng, et al. · CHI 2026 Chart data extraction, which reverse-engineers data tables from chart images, is essential for reproducibility, analysis, retrieval, and redesign. Existing interactive tools are reliable but tedious, and mixed-initiative systems, while more efficient, lack generalizability. Recent multimodal large language models (MLLM…https://arxiv.org/abs/2606.19782https://arxiv.org/abs/2606.19782Thu, 18 Jun 2026 00:00:00 +0000Aravind Narayanan, Shaina Raza Financial chart question answering in regulated settings demands more than accuracy: practitioners must know which answers to trust before acting on them, and many institutions cannot send client data to external model providers. Yet existing chart-QA agents are accuracy-focused and opaque, and most assume proprietary …https://arxiv.org/abs/2606.19646https://arxiv.org/abs/2606.19646Wed, 17 Jun 2026 00:00:00 +0000Ayush Dwivedi, Qixin Wang, Ashvi Soni, Ruoteng Wang, et al. Vision-language models (VLMs) are powerful for chart question answering, but invoking a VLM for every query can be unnecessarily expensive when many questions are answerable from OCR text and lightweight language reasoning. We demonstrate SAFE-Cascade, an interactive system for cost-adaptive chart question answering. G…https://arxiv.org/abs/2606.10640https://arxiv.org/abs/2606.10640Tue, 09 Jun 2026 00:00:00 +0000Hao Liu, Ruping Cao, Kun Wang, Zhiran Li, et al. In this report, we present our champion solution for the DataMFM Challenge Track 2: Chart Understanding. This track requires models to recover structured chart data and generate faithful natural-language summaries from chart images. To address the complementary requirements of accurate data extraction and factual narra…https://arxiv.org/abs/2606.01348https://arxiv.org/abs/2606.01348Sun, 31 May 2026 00:00:00 +0000Shangpin Peng, Gengluo Li, Xingyu Wan, Chengquan Zhang, et al. Charts are a primary medium for conveying quantitative and relational information, yet systematically evaluating chart parsing models remains difficult. Existing benchmarks focus on narrow chart types and leave diagrammatic structures such as flowcharts and mind maps largely unaddressed, while models produce outputs in…