# Robust Symbolic Reasoning for Visual Narratives via Hierarchical and Semantically Normalized Knowledge Graphs

arXiv: https://arxiv.org/abs/2508.14941 · HTML: https://mlsanity.com/paper/2508.14941/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 17 of its 40 references that are in the map.
- First posted on arXiv 20 Aug 2025.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 17 of 17)

- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [The NarrativeQA Reading Comprehension Challenge](https://mlsanity.com/paper/1712.07040/) (2017)
- [A Corpus and Evaluation Framework for Deeper Understanding of Commonsense Stories](https://mlsanity.com/paper/1604.01696/) (2016)
- [MovieQA: Understanding Stories in Movies through Question-Answering](https://mlsanity.com/paper/1512.02902/) (2015)
- [Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding](https://mlsanity.com/paper/1810.02338/) (2018)
- [Probing Pretrained Language Models for Lexical Semantics](https://mlsanity.com/paper/2010.05731/) (2020)
- [A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multimodal](https://mlsanity.com/paper/2212.05767/) (2022)
- [Fantastic Questions and Where to Find Them: FairytaleQA -- An Authentic Dataset for Narrative Comprehension](https://mlsanity.com/paper/2203.13947/) (2022)
- [Event Representations for Automated Story Generation with Deep Neural Nets](https://mlsanity.com/paper/1706.01331/) (2017)
- [DramaQA: Character-Centered Video Story Understanding with Hierarchical QA](https://mlsanity.com/paper/2005.03356/) (2020)
- [Knowledge-Enriched Visual Storytelling](https://mlsanity.com/paper/1912.01496/) (2019)
- [Large language models can segment narrative events similarly to humans](https://mlsanity.com/paper/2301.10297/) (2023)
- [Calibrating Concepts and Operations: Towards Symbolic Reasoning on Real Images](https://mlsanity.com/paper/2110.00519/) (2021)
- [A survey on knowledge-enhanced multimodal learning](https://mlsanity.com/paper/2211.12328/) (2022)
- [Collaborative Comic Generation: Integrating Visual Narrative Theories with AI Models for Enhanced Creativity](https://mlsanity.com/paper/2409.17263/) (2024)
- [Hierarchical Knowledge Graphs for Story Understanding in Visual Narratives](https://mlsanity.com/paper/2506.10008/) (2025)
- [A Customizable Generator for Comic-Style Visual Narrative](https://mlsanity.com/paper/2401.02863/) (2023)

## Built on by (most cited first, 1 of 1)

- [Narrative-to-Scene Generation: An LLM-Driven Pipeline for 2D Game Environments](https://mlsanity.com/paper/2509.04481/) (2025)

## Abstract (arXiv)

Understanding visual narratives such as comics requires structured representations that capture events, characters, and their relations across multiple levels of story organization. However, symbolic narrative graphs often suffer from inconsistency and redundancy, where similar actions or events are labeled differently across annotations or contexts. Such variance limits the effectiveness of reasoning and generalization. This paper introduces a semantic normalization framework for hierarchical narrative knowledge graphs. Building on cognitively grounded models of narrative comprehension, we propose methods that consolidate semantically related actions and events using lexical similarity and embedding-based clustering. The normalization process reduces annotation noise, aligns symbolic categories across narrative levels, and preserves interpretability. We demonstrate the framework on annotated manga stories from the Manga109 dataset, applying normalization to panel-, event-, and story-level graphs. Preliminary evaluations across narrative reasoning tasks, such as action retrieval, character grounding, and event summarization, show that semantic normalization improves coherence and robustness, while maintaining symbolic transparency. These findings suggest that normalization is a key step toward scalable, cognitively inspired graph models for multimodal narrative understanding.