# CoLa: Chinese Character Decomposition with Compositional Latent Components

arXiv: https://arxiv.org/abs/2506.03798 · HTML: https://mlsanity.com/paper/2506.03798/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 49 references that are in the map.
- First posted on arXiv 4 Jun 2025.

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Object-Centric Learning with Slot Attention](https://mlsanity.com/paper/2006.15055/) (2020)
- [Multi-Object Representation Learning with Iterative Variational Inference](https://mlsanity.com/paper/1903.00450/) (2019)
- [Bridging the Gap to Real-World Object-Centric Learning](https://mlsanity.com/paper/2209.14860/) (2022)
- [SPACE: Unsupervised Object-Oriented Scene Representation via Spatial Attention and Decomposition](https://mlsanity.com/paper/2001.02407/) (2020)
- [SAVi++: Towards End-to-End Object-Centric Learning from Real-World Videos](https://mlsanity.com/paper/2206.07764/) (2022)
- [Illiterate DALL-E Learns to Compose](https://mlsanity.com/paper/2110.11405/) (2021)
- [Simple Unsupervised Object-Centric Learning for Complex and Naturalistic Videos](https://mlsanity.com/paper/2205.14065/) (2022)
- [Neural Expectation Maximization](https://mlsanity.com/paper/1708.03498/) (2017)
- [Spatial Broadcast Decoder: A Simple Architecture for Learning Disentangled Representations in VAEs](https://mlsanity.com/paper/1901.07017/) (2019)
- [Generative Neurosymbolic Machines](https://mlsanity.com/paper/2010.12152/) (2020)
- [Efficient Iterative Amortized Inference for Learning Symmetric and Disentangled Multi-Object Representations](https://mlsanity.com/paper/2106.03630/) (2021)
- [Zero-Shot Chinese Character Recognition with Stroke-Level Decomposition](https://mlsanity.com/paper/2106.11613/) (2021)
- [Chinese Text Recognition with A Pre-Trained CLIP-Like Model Through Image-IDS Aligning](https://mlsanity.com/paper/2309.01083/) (2023)
- [DenseRAN for Offline Handwritten Chinese Character Recognition](https://mlsanity.com/paper/1808.04134/) (2018)
- [An open dataset for oracle bone script recognition and decipherment](https://mlsanity.com/paper/2401.15365/) (2024)
- [Multi-Column Deep Neural Networks for Offline Handwritten Chinese Character Classification](https://mlsanity.com/paper/1309.0261/) (2013)
- [Chinese Character Recognition with Radical-Structured Stroke Trees](https://mlsanity.com/paper/2211.13518/) (2022)
- [HierCode: A Lightweight Hierarchical Codebook for Zero-shot Chinese Text Recognition](https://mlsanity.com/paper/2403.13761/) (2024)

## Built on by (most cited first, 2 of 2)

- [Interpretable Oracle Bone Script Decipherment through Radical and Pictographic Analysis with LVLMs](https://mlsanity.com/paper/2508.10113/) (2025)
- [Towards Computational Chinese Paleography](https://mlsanity.com/paper/2601.06753/) (2026)

## Abstract (arXiv)

Humans can decompose Chinese characters into compositional components and recombine them to recognize unseen characters. This reflects two cognitive principles: Compositionality, the idea that complex concepts are built on simpler parts; and Learning-to-learn, the ability to learn strategies for decomposing and recombining components to form new concepts. These principles provide inductive biases that support efficient generalization. They are critical to Chinese character recognition (CCR) in solving the zero-shot problem, which results from the common long-tail distribution of Chinese character datasets. Existing methods have made substantial progress in modeling compositionality via predefined radical or stroke decomposition. However, they often ignore the learning-to-learn capability, limiting their ability to generalize beyond human-defined schemes. Inspired by these principles, we propose a deep latent variable model that learns Compositional Latent components of Chinese characters (CoLa) without relying on human-defined decomposition schemes. Recognition and matching can be performed by comparing compositional latent components in the latent space, enabling zero-shot character recognition. The experiments illustrate that CoLa outperforms previous methods in both character the radical zero-shot CCR. Visualization indicates that the learned components can reflect the structure of characters in an interpretable way. Moreover, despite being trained on historical documents, CoLa can analyze components of oracle bone characters, highlighting its cross-dataset generalization ability.