# LadderMoE: Ladder-Side Mixture of Experts Adapters for Bronze Inscription Recognition

arXiv: https://arxiv.org/abs/2510.01651 · HTML: https://mlsanity.com/paper/2510.01651/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 40 references that are in the map.
- First posted on arXiv 2 Oct 2025.

## Builds on (most cited first, 25 of 25)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Mixtral of Experts](https://mlsanity.com/paper/2401.04088/) (2024)
- [Mixture-of-Experts with Expert Choice Routing](https://mlsanity.com/paper/2202.09368/) (2022)
- [An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition](https://mlsanity.com/paper/1507.05717/) (2015)
- [YOLOv12: Attention-Centric Real-Time Object Detectors](https://mlsanity.com/paper/2502.12524/) (2025)
- [Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition](https://mlsanity.com/paper/2103.06495/) (2021)
- [What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis](https://mlsanity.com/paper/1904.01906/) (2019)
- [Scene Text Recognition with Permuted Autoregressive Sequence Models](https://mlsanity.com/paper/2207.06966/) (2022)
- [Focusing Attention: Towards Accurate Text Recognition in Natural Images](https://mlsanity.com/paper/1709.02054/) (2017)
- [Vision Transformer for Fast and Efficient Scene Text Recognition](https://mlsanity.com/paper/2105.08582/) (2021)
- [Sequence-to-Sequence Contrastive Learning for Text Recognition](https://mlsanity.com/paper/2012.10873/) (2020)
- [Rosetta: Large scale system for text detection and recognition in images](https://mlsanity.com/paper/1910.05085/) (2019)
- [TextScanner: Reading Characters in Order for Robust Scene Text Recognition](https://mlsanity.com/paper/1912.12422/) (2019)
- [What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels](https://mlsanity.com/paper/2103.04400/) (2021)
- [Scene Text Recognition from Two-Dimensional Perspective](https://mlsanity.com/paper/1809.06508/) (2018)
- [Parameter-Efficient Fine-Tuning in Large Models: A Survey of Methodologies](https://mlsanity.com/paper/2410.19878/) (2024)
- [Reading Scene Text in Deep Convolutional Sequences](https://mlsanity.com/paper/1506.04395/) (2015)
- [CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model](https://mlsanity.com/paper/2305.14014/) (2023)
- [SimAN: Exploring Self-Supervised Representation Learning of Scene Text via Similarity-Aware Normalization](https://mlsanity.com/paper/2203.10492/) (2022)
- [Unchosen Experts Can Contribute Too: Unleashing MoE Models' Power by Self-Contrast](https://mlsanity.com/paper/2405.14507/) (2024)
- [Symmetrical Linguistic Feature Distillation with CLIP for Scene Text Recognition](https://mlsanity.com/paper/2310.04999/) (2023)
- [Towards the Unseen: Iterative Text Recognition by Distilling from Errors](https://mlsanity.com/paper/2107.12081/) (2021)
- [Clustering-based Feature Representation Learning for Oracle Bone Inscriptions Detection](https://mlsanity.com/paper/2508.18641/) (2025)

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

- [Chronicles-OCR: A Cross-Temporal Perception Benchmark for the Evolutionary Trajectory of Chinese Characters](https://mlsanity.com/paper/2605.11960/) (2026)

## Abstract (arXiv)

Bronze inscriptions (BI), engraved on ritual vessels, constitute a crucial stage of early Chinese writing and provide indispensable evidence for archaeological and historical studies. However, automatic BI recognition remains difficult due to severe visual degradation, multi-domain variability across photographs, rubbings, and tracings, and an extremely long-tailed character distribution. To address these challenges, we curate a large-scale BI dataset comprising 22454 full-page images and 198598 annotated characters spanning 6658 unique categories, enabling robust cross-domain evaluation. Building on this resource, we develop a two-stage detection-recognition pipeline that first localizes inscriptions and then transcribes individual characters. To handle heterogeneous domains and rare classes, we equip the pipeline with LadderMoE, which augments a pretrained CLIP encoder with ladder-style MoE adapters, enabling dynamic expert specialization and stronger robustness. Comprehensive experiments on single-character and full-page recognition tasks demonstrate that our method substantially outperforms state-of-the-art scene text recognition baselines, achieving superior accuracy across head, mid, and tail categories as well as all acquisition modalities. These results establish a strong foundation for bronze inscription recognition and downstream archaeological analysis.