DeepVision-103K: A Visually Diverse, Broad-Coverage, and Verifiable Mathematical Dataset for Multimodal Reasoning

Haoxiang Sun; Lizhen Xu; Bing Zhao; Wotao Yin; Wei Wang; Boyu Yang; Rui Wang; Hu Wei

cs.LGFeb 18, 2026

DeepVision-103K: A Visually Diverse, Broad-Coverage, and Verifiable Mathematical Dataset for Multimodal Reasoning

Authors: Haoxiang Sun, Lizhen Xu, Bing Zhao, Wotao Yin, Wei Wang, Boyu Yang, Rui Wang, Hu Wei

Organizations: Alibaba Group · Shanghai Jiao Tong University

Abstract

Reinforcement Learning with Verifiable Rewards (RLVR) has been shown effective in enhancing the visual reflection and reasoning capabilities of Large Multimodal Models (LMMs). However, existing datasets are predominantly derived from either small-scale manual construction or recombination of prior resources, which limits data diversity and coverage, thereby constraining further gains in model performance. To this end, we introduce DeepVision-103K, a comprehensive dataset for RLVR training that covers diverse K12 mathematical topics, extensive knowledge points, and rich visual elements. Models trained on DeepVision achieve strong performance on multimodal mathematical benchmarks, and generalize effectively to general multimodal reasoning tasks. Further analysis reveals enhanced visual perception, reflection and reasoning capabilities in trained models, validating DeepVision's effectiveness for advancing multimodal reasoning. Data: this url.

Figures & tables

Figure 1: The number of different visual element types of training datasets.

Figure 2: Performance on multimodal math and general multimodal benchmarks, we report averaged Pass@1 accuracy across benchmarks.

Figure 3: A data sample from DeepVision-103K.

Field	Description
Question & Image	A multimodal mathematical problem consisting of a textual problem statement and the corresponding image.
Final Answer	A unique, verifiable answer that enables rule-based reward computation in RLVR.
Pass Rate	The proportion of correct responses obtained during model rollouts.
Topic	A hierarchical classification indicating which branch of mathematics the problem belongs to.
Knowledge Points	A list of specific mathematical concepts, theorems, or techniques required to solve the problem.
Visual Elements	A list of geometric or graphical objects depicted in the image, describing what visual content should be perceived and interpreted.

Table 1: Annotation fields and definitions.

Figure 4: Visual elements in DeepVision-103K.

Category	Key Visual Elements
Planar Geometry	Primitives (Angle, Triangle, Circle, Quadrilateral,Polygon), Relations (Parallelism, Tangency, Chords), Properties (Right Angles, Perpendicularity)
Solid Geometry	3D Primitives (Cube, Prism, Cylinder, Cone), Spatial Representations (Orthographic Views, Nets), Sections (Frustums, Hemispheres)
Analytic Plot	Coordinate Systems, Function Curves (Linear, General), Conic Sections (Parabola, Hyperbola), Scatter Points, Inequality Regions
Data Chart	Statistical Graphs (Bar, Histogram, Pie, Line), Structured Data (Tables, Stem-and-Leaf)
Schematic Diagram	Logical Structures (Flowcharts, Tree Diagrams), Physics/Sets (Force Diagrams, Circuits, Venn Diagrams), Linear Arrangements
Real-World Item	Objects (Characters, Household Items), Contextual Scenes (Architecture, Maps, Scientific Tools)

Table 2: Visual categories and element coverage in DeepVision-103K.

Figure 5: Mathematical topics in DeepVision-103K.

Figure 6: Curation pipeline for mathematical data in DeepVision-103K.

Model	Multimodal Math				General Multimodal
Model	WeMath	MathVision	$\textbf{MathVerse}_{\textbf{vision}}$	LogicVista	$\textbf{MMMU}_{\textbf{val}}$	$\textbf{MMMU}_{\textbf{Pro}}$	$\textbf{M}^{3}\textbf{CoT}$
Closed-source Models
GPT-5-Nano-High	78.62	58.75	70.30	58.03	70.78	70.64	69.15
Gemini-2.5-Flash-Lite	83.85	52.47	70.30	60.49	64.77	65.08	68.42
Qwen3-VL-8B Series
Qwen3-VL-8B-Instruct	79.36	51.44	67.38	61.16	67.66	67.69	70.83

Table 3: Performance comparison across multimodal mathematical reasoning and general multimodal benchmarks. We report Pass@1 accuracy (%). The best results for each model family are shown in bold .

Figure 8: DeepVision model correctly identifies the shaded region on the first attempt.

Figure 9: DeepVision model actively re-examines visual content to correct errors, while the base model merely rephrases without genuine verification.

Figure 10: DeepVision model systematically enumerates all possible angle combinations and concludes the type cannot be determined, while the Instruct model incorrectly assumes symmetry without justification.

Data Composition	Multimodal Math					General Multimodal
Data Composition	WeMath	MathVision	MathVerse	LogicVista	Avg.	MMMU ${}_{\text{val}}$	MMMU ${}_{\text{pro}}$	M3CoT	Avg.
MiMo-VL-7B	74.42	50.69	72.71	60.71	64.63	63.77	60.69	70.02	64.83
DeepVision-103K 200	82.98	55.23	76.26	65.92	70.10	71.00	69.19	72.56	70.92
w/o visual logic data
Math-77K 150	81.67	54.83	74.23	63.98	68.68	70.00	68.55	72.09	70.21
Math-77K 200	82.07	55.72	74.74	63.53	69.02	68.50	69.67	72.65	70.27

Table 4: Ablation studies on data composition and quality. We report Pass@1 accuracy (%) across mathematical reasoning and general multimodal benchmarks. All experiments used MiMo-VL-7B-SFT-2508 as the base model.

Appendix figures & tables13 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 11: Solid Geometry & Analytic Plots.

Figure 12: Planar Geometry & Solid Geometry & Real-World Item.

Category	Fine-grained types
planar_geometry	Right Triangle; Equilateral Triangle; Triangle; Square; Rectangle; Rhombus; Parallelogram; Trapezoid; Quadrilateral; Circle; Semicircle; Sector; Arc; Parallel Lines; Perpendicular Lines; Tangent; Chord; Angle; Right Angle
solid_geometry	Cube; Cuboid; Prism; Pyramid; Tetrahedron; Sphere; Cylinder; Cone; Frustum; Hemisphere; Net; Orthographic View
analytic_plot	Linear Graph; Parabola; Hyperbola; Sinusoidal Curve; Exponential Curve; Analytic Circle; General Function Curve; Coordinate System; Number Line; Scatter Points; Inequality Region; Equation
data_chart	Table; Bar Chart; Line Chart; Pie Chart; Donut Chart; Histogram; Box Plot; Stem-and-Leaf Plot
schematic_diagram	Flowchart; Circuit; Force Diagram; Tree Diagram; Venn Diagram; Linear Arrangement
real-world item	Character; Plant; Scientific Tool; Vehicle; Architecture; Household Item; Apparel; Food; Real Object; Scene; Map

Appendix

Table 5: Visual-element annotation taxonomy used in this work.

Figure 13: Top 10 Knowledge-based retrieval. The x-axis IDs correspond to knowledge domains listed in Table 6 .

ID	Knowledge Domain	w/o Retrieval	w/ Retrieval	Increase
1	Circle $\rightarrow$ Inscribed and Circumscribed	771	2,079	+1,308
2	Triangle $\rightarrow$ Angle of Elevation and Depression	765	1,654	+889
3	Circle $\rightarrow$ Tangency	624	1,384	+760
4	Circle $\rightarrow$ Perpendicular Chord Theorem	284	697	+413
5	Conic Sections $\rightarrow$ Hyperbola	159	410	+251
6	Figure Relationships $\rightarrow$ Inscribed and Circumscribed	215	447	+232

Appendix

Table 6: Top 10 Knowledge Domains by Retrieval Gap

Label	Category	Trigger
ERR_IMG_MISSING	Image quality issue	Image is missing, unreadable, or lacks essential visual information.
ERR_TEXT_MISSING	Missing text	Question text misses key conditions/values, making the task unsolvable.
ERR_MISMATCH	Image–text mismatch	Image content conflicts with the question statement.
CORRECTION	Incorrect reference answer	Data are valid, but the reference answer is incorrect; return the corrected solution/answer in L a T e X .
1	Perfect match	Image/text are complete and consistent, and the reference answer is correct.

Appendix

Table 7: Verification labels used by Gemini 3 Flash . Exactly one label is returned per instance.

Data Source	License	URL
MM-MathInstruct-3M ( Wang et al., 2025c )	Apache 2.0	https://huggingface.co/datasets/MathLLMs/MM-MathInstruct
MultiMath-300K ( Peng et al., 2024 )	Unset	https://huggingface.co/datasets/pengshuai-rin/multimath-300k
Zebra-CoT ( Li et al., 2025 )	CC BY-NC 4.0	https://huggingface.co/datasets/multimodal-reasoning-lab/Zebra-CoT
GameQA-140K Tong et al. (2025)	MIT	https://huggingface.co/datasets/Code2Logic/GameQA-140K
PuzzleVQA Chia et al. (2024)	Unset	https://huggingface.co/datasets/declare-lab/PuzzleVQA

Appendix

Table 8: Licenses and usage permissions for the data sources used in this work.

Config	Value
lr	1e-6
kl_coef	1e-3
max_prompt_length	2K
max_response_length	16K
gen_batch_size	512
train_batch_size	256

Appendix

Table 9: Configurations for training DeepVision series models.

Category	Benchmark	#Samples	Reference
Multimodal Math	WeMath	1,740	( Qiao et al., 2024 )
	MathVision	3,040	( Wang et al., 2024 )
	$\text{MathVerse}_{\text{vision}}$	788	( Zhang et al., 2024 )
	LogicVista	448	( Xiao et al., 2024 )
General Multimodal	$\text{M}^{3}\text{CoT}$	2,318	( Chen et al., 2024 )
	$\text{MMMU}_{\text{val}}$	900	( Yue et al., 2024a )

Appendix

Table 10: Overview of evaluation benchmarks.

Parameter	Qwen3-VL-Thinking	Qwen3-VL-Instruct	MiMo-VL-(SFT/RL)
top_p	0.95	0.8	0.95
top_k	20	20	–
temperature	1.0	0.7	0.3
repetition_penalty	1.0	1.0	–
presence_penalty	0.0	1.5	–
max_tokens	32,768	32,768	32,768

Appendix

Table 11: Inference hyperparameters for each model family.

Multimodal Reasoning

Multimodal reasoning studies how AI systems combine text with images, audio, video, and other signals to solve evidence-dependent tasks.

See all 508 papers

UniCAR-RL: Seeing Better before Thinking Deeper in Visual Mathematics
Sep 12, 2026Yuzhe Li, Hao Yan, Hao Wang +6VLM Reasoning Large Vision-Language Models
Huazhong University of Science and Technology · Huawei Inc.
Multimodal Mathematical Reasoning with Diverse Solving Perspective
Jul 3, 2025Wenhao Shi, Zhiqiang Hu, Yi Bin +4Multimodal Reasoning Multimodal CoT Reasoning
Tongji University, Shanghai, China · National University of Singapore, Singapore · Tencent, Beijing, China +2
MathVis-Fine: Aligning Visual Supervision with Necessity via Progressive Dependency-Guided Training for Multimodal Mathematical Reasoning
Jun 16, 2026Wanshi Xu, Haokun Zhao, Haidong Yuan +2Multimodal Reasoning Multimodal Reward Modeling
School of ECE, Peking University · College of Computer Science and Artificial Intelligence, Fudan University · School of Software and Microelectronics, Peking University +1

Sep 12, 2026J/K move · Enter open · S save

UniCAR-RL: Seeing Better before Thinking Deeper in Visual Mathematics

Yuzhe Li, Hao Yan, Hao Wang, Xingchen Liu+5

Huazhong University of Science and Technology · Huawei Inc.

VLM Reasoning Large Vision-Language Models Mathematical Reasoning Visually Grounded Reasoning RL for VLMs

Open paper AskPDF arXiv

Multimodal Large Language Models (MLLMs) often struggle with complex mathematical visual reasoning primarily due to a lack of fine-grained perception, causing initial visual hallucinations to directly trigger cascading reasoning failures. In traditional end-to-end reinforcement learning (RL), sparse rewards fail to decouple perceptual hallucinations from logical missteps, hindering targeted perception optimization. Alternatively, fine-tuning with perception-enhanced CoT data incurs high costs and hallucinations. In this paper, we address these challenges by proposing UniCAR-RL, an annotation-free RL framework. By explicitly decoupling the optimization of perception and reasoning during the training process, it achieves isolation and optimization of both capabilities. Specifically, UniCAR-RL consists of three synergistic branches: 1) a Caption-RL branch that optimizes perception capabilities through verifier-guided reasoning validation; 2) a Reasoning-RL branch that performs logical reasoning based on a gold image description to halt cascading errors; 3) a QA-RL branch that retains native end-to-end alignment to ensure robust question-answering performance. Experiments show that UniCAR-RL substantially improves MLLMs' mathematical and visual reasoning using only raw short-answer data. Furthermore, it demonstrates strong generalization across diverse architectures and scales.

Abstract

Multimodal Reasoning

Explore similar work

UniCAR-RL: Seeing Better before Thinking Deeper in Visual Mathematics

Multimodal Mathematical Reasoning with Diverse Solving Perspective

MathVis-Fine: Aligning Visual Supervision with Necessity via Progressive Dependency-Guided Training for Multimodal Mathematical Reasoning