# Geoparsing: Diagram Parsing for Plane and Solid Geometry with a Unified Formal Language

Source: [Geoparsing: Diagram Parsing for Plane and Solid Geometry with a Unified Formal Language](<https://www.sota2.com/research/paper/69ddca7c7d0172140ec79f7f>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-18T05:00:04.772Z

Authors: Peijie Wang, Ming-Liang Zhang, Jun Cao, Chao Deng, Dekang Ran, Hongda Sun, Pi Bu, Xuan Zhang, Yingyao Wang, Jun Song, Bo Zheng, Fei Yin, Cheng-Lin Liu

Published: 2026-04-13T15:09:56.000Z

## Abstract

Multimodal Large Language Models (MLLMs) have achieved remarkable progress but continue to struggle with geometric reasoning, primarily due to the perception bottleneck regarding fine-grained visual elements. While formal languages have aided plane geometry understanding, solid geometry which requires spatial understanding remains largely unexplored. In this paper, we address this challenge by designing a unified formal language that integrates plane and solid geometry, comprehensively covering geometric structures and semantic relations. We construct GDP-29K, a large-scale dataset comprising 20k plane and 9k solid geometry samples collected from diverse real-world sources, each paired with its ground-truth formal description. To ensure syntactic correctness and geometric consistency, we propose a training paradigm that combines Supervised Fine-Tuning with Reinforcement Learning via Verifiable Rewards. Experiments show that our approach achieves state-of-the-art parsing performance. Furthermore, we demonstrate that our parsed formal descriptions serve as a critical cognitive scaffold, significantly boosting MLLMs' capabilities for downstream geometry reasoning tasks. Our data and code are available at Geoparsing.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Geometric diagram parsing on SGDP-1K (test)](<https://www.sota2.com/research/sota/geometric-diagram-parsing-on-sgdp-1k-test.md>): Geometric diagram parsing; SGDP-1K (test); Point Precision: 99.2.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.11600>)

[PDF](<https://arxiv.org/pdf/2604.11600.pdf>)
