OracleFusion: Assisting the Decipherment of Oracle Bone Script with Structurally Constrained Semantic Typography

2025-06-26

AI Summary

Approximately 4,500 oracle bone inscriptions (OBIs) have been unearthed, yet only ~1,600 characters have been deciphered; the remaining undeciphered characters pose significant challenges due to their complex structures and abstract semantic imagery. To address this, we propose OracleFusion—a two-stage semantic-graphic fusion framework that uniquely integrates a spatially aware reasoning multimodal large language model (MLLM-SAR) with an oracle-specific vector fusion method (OSVF), enabling high-fidelity, structure-constrained vector font generation. OracleFusion precisely localizes critical graphical components and injects semantic information while preserving original structural integrity, thereby markedly enhancing legibility and visual expressiveness for unseen characters. Experiments demonstrate that OracleFusion outperforms existing approaches in semantic accuracy, visual quality, and glyph fidelity. It provides archaeolinguists with reliable decipherment cues and substantially improves the efficiency of OBI interpretation.

Technology Category

• Computer Vision: Language and Vision
• Natural Language Processing: Lexical Semantics and Morphology
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

As one of the earliest ancient languages, Oracle Bone Script (OBS) encapsulates the cultural records and intellectual expressions of ancient civilizations. Despite the discovery of approximately 4,500 OBS characters, only about 1,600 have been deciphered. The remaining undeciphered ones, with their complex structure and abstract imagery, pose significant challenges for interpretation. To address these challenges, this paper proposes a novel two-stage semantic typography framework, named OracleFusion. In the first stage, this approach leverages the Multimodal Large Language Model (MLLM) with enhanced Spatial Awareness Reasoning (SAR) to analyze the glyph structure of the OBS character and perform visual localization of key components. In the second stage, we introduce Oracle Structural Vector Fusion (OSVF), incorporating glyph structure constraints and glyph maintenance constraints to ensure the accurate generation of semantically enriched vector fonts. This approach preserves the objective integrity of the glyph structure, offering visually enhanced representations that assist experts in deciphering OBS. Extensive qualitative and quantitative experiments demonstrate that OracleFusion outperforms state-of-the-art baseline models in terms of semantics, visual appeal, and glyph maintenance, significantly enhancing both readability and aesthetic quality. Furthermore, OracleFusion provides expert-like insights on unseen oracle characters, making it a valuable tool for advancing the decipherment of OBS.

Problem

• Deciphering undeciphered Oracle Bone Script characters with complex structures
• Generating semantically enriched vector fonts for ancient script analysis
• Enhancing readability and aesthetics of Oracle Bone Script interpretations

Innovation

• Two-stage semantic typography framework OracleFusion
• MLLM with Spatial Awareness Reasoning (SAR)
• Oracle Structural Vector Fusion (OSVF)