# MathSticks: A Benchmark for Visual Symbolic Compositional Reasoning with Matchstick Puzzles

arXiv: https://arxiv.org/abs/2510.00483 · HTML: https://mlsanity.com/paper/2510.00483/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal reasoning, one of 17 directions in LLM reasoning.
- Builds on 29 of its 64 references that are in the map.
- First posted on arXiv 1 Oct 2025.

## Placed in

- LLM reasoning / [Multimodal reasoning](https://mlsanity.com/lr/multimodal-reasoning/)

## Builds on (most cited first, 29 of 29)

- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [MathVista: Evaluating Mathematical Reasoning of Foundation Models in Visual Contexts](https://mlsanity.com/paper/2310.02255/) (2023)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [WorldSimBench: Towards Video Generation Models as World Simulators](https://mlsanity.com/paper/2410.18072/) (2024)
- [HybridVLA: Collaborative Diffusion and Autoregression in a Unified Vision-Language-Action Model](https://mlsanity.com/paper/2503.10631/) (2025)
- [Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning](https://mlsanity.com/paper/2504.13914/) (2025)
- [RoboRefer: Towards Spatial Referring with Reasoning in Vision-Language Models for Robotics](https://mlsanity.com/paper/2506.04308/) (2025)
- [Draw-and-Understand: Leveraging Visual Prompts to Enable MLLMs to Comprehend What You Want](https://mlsanity.com/paper/2403.20271/) (2024)
- [Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning of Vision Language Models](https://mlsanity.com/paper/2503.20752/) (2025)
- [Beyond Attention or Similarity: Maximizing Conditional Diversity for Token Pruning in MLLMs](https://mlsanity.com/paper/2506.10967/) (2025)
- [RoboBrain 2.0 Technical Report](https://mlsanity.com/paper/2507.02029/) (2025)
- [Code-as-Monitor: Constraint-aware Visual Programming for Reactive and Proactive Robotic Failure Detection](https://mlsanity.com/paper/2412.04455/) (2024)
- [LaMP: Language-Motion Pretraining for Motion Generation, Retrieval, and Captioning](https://mlsanity.com/paper/2410.07093/) (2024)
- [MineDreamer: Learning to Follow Instructions via Chain-of-Imagination for Simulated-World Control](https://mlsanity.com/paper/2403.12037/) (2024)
- [RoboOS: A Hierarchical Embodied Framework for Cross-Embodiment and Multi-Agent Collaboration](https://mlsanity.com/paper/2505.03673/) (2025)
- [EmbodiedOcc++: Boosting Embodied 3D Occupancy Prediction with Plane Regularization and Uncertainty Sampler](https://mlsanity.com/paper/2504.09540/) (2025)
- [ManipLVM-R1: Reinforcement Learning for Reasoning in Embodied Manipulation with Large Vision-Language Models](https://mlsanity.com/paper/2505.16517/) (2025)
- [RH20T-P: A Primitive-Level Robotic Dataset Towards Composable Generalization Agents](https://mlsanity.com/paper/2403.19622/) (2024)
- [MSC-Bench: Benchmarking and Analyzing Multi-Sensor Corruption for Driving Perception](https://mlsanity.com/paper/2501.01037/) (2025)
- [VisualTrans: A Benchmark for Real-World Visual Transformation Reasoning](https://mlsanity.com/paper/2508.04043/) (2025)
- [ManipDreamer3D : Synthesizing Plausible Robotic Manipulation Video with Occupancy-aware 3D Trajectory](https://mlsanity.com/paper/2509.05314/) (2025)
- [General Point Model with Autoencoding and Autoregressive](https://mlsanity.com/paper/2310.16861/) (2023)
- [What Really Matters for Robust Multi-Sensor HD Map Construction?](https://mlsanity.com/paper/2507.01484/) (2025)
- [EgoPrompt: Prompt Learning for Egocentric Action Recognition](https://mlsanity.com/paper/2508.03266/) (2025)
- [Visual Transformation Telling](https://mlsanity.com/paper/2305.01928/) (2023)
- [FastRSR: Efficient and Accurate Road Surface Reconstruction from Bird's Eye View](https://mlsanity.com/paper/2504.09535/) (2025)
- [MulSMo: Multimodal Stylized Motion Generation by Bidirectional Control Flow](https://mlsanity.com/paper/2412.09901/) (2024)

## Built on by (most cited first, 5 of 5)

- [TIGeR: Tool-Integrated Geometric Reasoning in Vision-Language Models for Robotics](https://mlsanity.com/paper/2510.07181/) (2025)
- [Towards Cross-View Point Correspondence in Vision-Language Models](https://mlsanity.com/paper/2512.04686/) (2025)
- [Scaling Up AI-Generated Image Detection with Generator-Aware Prototypes](https://mlsanity.com/paper/2512.12982/) (2025)
- [PReD: An LLM-based Foundation Multimodal Model for Electromagnetic Perception, Recognition, and Decision](https://mlsanity.com/paper/2603.28183/) (2026)
- [SARFusion: Scene-Aware Routing Fusion for Robust Camera-LiDAR 3D Object Detection](https://mlsanity.com/paper/2609.29235/) (2026)

## Abstract (arXiv)

We introduce textscMathSticks, a benchmark for Visual Symbolic Compositional Reasoning (VSCR), which unifies visual perception, symbolic manipulation, and arithmetic consistency. Each task presents an incorrect matchstick equation that must be corrected by moving one or two sticks under strict conservation rules. The benchmark includes both text-guided and purely visual settings, systematically covering digit scale, move complexity, solution multiplicity, and operator variation, with 1.4M generated instances and a curated test set. Evaluations of 14 vision--language models reveal substantial limitations: closed-source models succeed only on simple cases, open-source models fail in the visual regime, while humans exceed 90% accuracy. These findings establish textscMathSticks as a rigorous testbed for advancing compositional reasoning across vision and symbols. Our code and dataset are publicly available at https://github.com/Yuheng2000/MathSticks.