SGA: Plug&Play Geometric Verification for Educational Video Synthesis

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Lopez Jhon, Hinojosa Carlos, Ghanem Bernard
• Venue: arXiv: cs.AI
• Published: 2026-07-20
• Collected: 2026-07-21
• URL: http://arxiv.org/abs/2607.18116v1
• PDF: https://arxiv.org/pdf/2607.18116v1
• DOI: UNKNOWN

Plain-Language Brief

Researchers propose a new module called Symbolic Geometric Agent (SGA) that works with existing animation pipelines to improve spatial correctness and visual legibility for educational videos. The SGA intercepts code generated by large language models, extracts symbolic scene graphs, and refines the code when spatial conflicts are detected. They also introduce a new scoring system called Manim Visual Quality Score (MVQS) that measures spatial integrity without rendering the videos. Experiments show that using the SGA leads to improvements in MVQS for various large language models and animation pipelines.

What the Document Says

• The paper presents the Symbolic Geometric Agent (SGA), a module designed to improve spatial correctness and visual legibility for educational videos.
• SGA works with existing code-centric animation pipelines that use large language models like Manim.
• SGA intercepts the code generated by these large language models, performs partial execution, and extracts symbolic scene graphs.
• When spatial conflicts are detected in the scene graphs, SGA applies targeted refinement to improve accuracy.
• The researchers also introduce the Manim Visual Quality Score (MVQS), a measure of spatial integrity that doesn't require rendering videos.
• Experiments were conducted on the MMMC-Code benchmark using four different large language model backbones and two agentic pipelines.
• The results show that SGA achieves a peak MVQS of 73.11 with the Code2Video + GPT-5.1 combination, representing a 16.1% improvement over the baseline.
• In 7 out of 8 combinations of backbones and pipelines, SGA improved the MVQS.

Key Concepts Mentioned

• Symbolic Geometric Agent (SGA)
• Manim
• Large Language Models (LLMs)
• Manim Visual Quality Score (MVQS)
• spatial correctness
• visual legibility
• educational videos

Claims Made by the Document

• SGA improves spatial correctness and visual legibility for educational videos.
• SGA works with existing animation pipelines that use large language models like Manim.

Evidence Used

• The paper describes the development and implementation of SGA.
• Results from experiments on the MMMC-Code benchmark are presented.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Recent work leverages Large Language Models (LLMs) to generate executable code for pedagogical animations using libraries such as Manim. However, ensuring spatial correctness and visual legibility remains challenging, as existing frameworks emphasize pedagogical content while overlooking geometric occlusions. We propose the Symbolic Geometric Agent (SGA), a plug-and-play module for code-centric animation pipelines that intercepts LLM-generated code, performs partial execution to extract symbolic scene graphs, and applies targeted refinement when spatial conflicts are detected. We further introduce the Manim Visual Quality Score (MVQS), a deterministic rendering-free proxy for spatial integrity. Experiments on the MMMC-Code benchmark across four LLM backbones and two agentic pipelines show that SGA achieves a peak MVQS of 73.11 (Code2Video + GPT-5.1), corresponding to a 16.1% relative improvement over the raw baseline, and improves MVQS in 7 of 8 backbone x pipeline configurations.