CURV: Enhancing Chart Understanding Through Curriculum Visual Grounded Reasoning
Summary
CURV proposes a curriculum learning framework that enhances chart question answering by developing intrinsic visual grounded reasoning in multimodal LLMs, achieving significant improvements on real-world benchmarks.
View Cached Full Text
Cached at: 08/05/26, 05:46 PM
Paper page - CURV: Enhancing Chart Understanding Through Curriculum Visual Grounded Reasoning
Source: https://huggingface.co/papers/2608.02833
Abstract
Chartquestionanswering(CQA)requiresmultimodallargelanguagemodels(MLLMs)tointegratevisualcomprehensionwithlogicalreasoning,yetcurrentmodelsstrugglewithaccuratevisualgroundingandcoherentreasoningchains.Whileextrinsicchain-of-thoughtpromptingandvisualcuessignificantlyimproveperformance,currentMLLMslackintrinsicvisualgroundedreasoningcapabilities,leadingtoinaccurateperceptionandreasoningdisconnectedfromvisualevidence.Toaddresstheselimitations,weproposeCURV,acurriculumlearningframeworkthatdevelopsintrinsicvisualreasoningcapabilitiesbyreformulatingCQAasmulti-stepvisualgroundedreasoning,whereeachstepcoordinateslogicalreasoningwithdynamicvisualgroundingthroughspatialattentionconcentration.Toassistmodellearning,wefurtherintroduceCCQA,athree-levelcurriculumdatasetwithscalablesyntheticgenerationacrossdiversecharttypesandreasoningpatterns.Ourcurriculumsystematicallyprogressesfrombasicsingle-operationreasoningtocomplexmulti-chartcompositionaltasks.ExperimentsdemonstratethatCURVachievesuptouparrow20.50%improvementsoverbaselinesandisgeneralizabletoreal-worldbenchmarks(uptouparrow12.30%)andout-of-domainmultimodalreasoningtasks(uptouparrow10.20%),validatingtheeffectivenessofinternalizingvisualreasoningwithdynamicgroundingforenhancedchartunderstandingcapabilities.Codeisavailableat:https://xhguo7.github.io/CURV/.
View arXiv pageView PDFProject pageGitHub0Add to collection
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2608.02833 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2608.02833 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2608.02833 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
Navigating the Mirage: A Dual-Path Agentic Framework for Robust Misleading Chart Question Answering
This paper introduces ChartCynics, an agentic dual-path framework that decouples perception from verification to robustly answer questions about misleading charts. It achieves state-of-the-art accuracy by using a diagnostic vision path and an OCR-driven data path, with a two-stage protocol for reasoning distillation and adversarial alignment.
Unlocking Complex Visual Generation via Closed-Loop Verified Reasoning
Introduces CLVR (Closed-Loop Visual Reasoning), a framework that reformulates text-to-image generation from a single-step process into a closed-loop, multi-step visual reasoning approach using a VLM controller and diffusion models, achieving improved performance on compositional prompts.
Anchoring What Matters: A Dual-Level Learning Framework for Visually-Grounded Multimodal Reasoning
This paper proposes PIVOT, a dual-level learning framework that enhances visually-grounded reasoning in large vision-language models by using self-calibrated experience replay and vision-guided advantage allocation to optimize reinforcement learning.
CVPO: Enhancing LLM Reinforcement Learning Reasoning via Value-Variance Adaptation and Dynamic Curriculum Learning
This arXiv paper introduces CVPO, a reinforcement learning method for LLMs that adapts value-variance for advantage estimation and uses dynamic curriculum learning to match question difficulty, achieving better reasoning performance than VAPO on math tasks.
Reason Through the Latent! Making Latent Visual Reasoning Necessary
The paper introduces Causal Visual Recurrent Reasoning (CVRR), a method that enforces recurrent hidden-state computation for visual reasoning, improving performance on benchmarks while distinguishing latent informativeness from actual predictive use.