Reproduce “Towards Computational Chinese Paleography” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yiran Rex Ma
“Towards Computational Chinese Paleography” frames computational paleography as more than digitization or character recognition: it is a research discipline spanning image restoration, contextual analysis, decipherment, and historical interpretation. Through a literature-based survey, it organizes datasets and methods into a full pipeline, identifies persistent weaknesses in contextual and multimodal reasoning, and advocates human-AI collaboration supported by richer, better-balanced resources.
Key insights
- Computational Chinese paleography should participate in core scholarly reasoning—including collation, dating, decipherment, and synthesis—not merely provide digitized archives or recognition tools.
- Radical-aware, compositional, few-shot, and open-set methods are especially important because ancient Chinese characters have many variants, rare classes, and previously unseen or undeciphered forms.
- The field needs more contextual and multimodal data: isolated glyph images are abundant, but complete artifacts, document layouts, textual sequences, archaeological metadata, and semantic links remain underrepresented.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.