---
license: apache-2.0
pipeline_tag: feature-extraction
library_name: transformers
---

# LCO-Embedding: Scaling Language-Centric Omnimodal Representation Learning

We are thrilled to release LCO-Embedding - a language-centric omnimodal representation learning framework and the LCO-Embedding model families!

This model implements the framework presented in the paper [Scaling Language-Centric Omnimodal Representation Learning](https://huggingface.co/papers/2510.11693), accepted by NeurIPS 2025.

**Project Page:** https://huggingface.co/LCO-Embedding
**Github Repository:** https://github.com/LCO-Embedding/LCO-Embedding

## Overview

We introduce **LCO-Embedding**, a language-centric omnimodal representation learning method and the LCO-Embedding model families, setting a new state-of-the-art on [MIEB](https://huggingface.co/blog/isaacchung/introducing-mieb) (Massive Image Embedding Benchmark), while supporting audio and videos.

This work also introduces the **Generation-Representation Scaling Law**, connecting models' generative capabilities and their representation upper bound. Furthermore, we introduce **SeaDoc**, a challenging visual document retrieval task in Southeast Asian languages, and show that continual generative pretraining before contrastive learning raises the representation upper bound.

<div align='center'><img src="https://cdn-uploads.huggingface.co/production/uploads/604f67ef0fe8ff3ec13d71ef/4Wd8fDFBdT6GxqN6-KzZN.png" alt="overview" width="100%"/></div>

## Evaluation Results

We evaluate LCO-Embedding with state-of-the-art embedding models, including E5-V, Voyage Multimodal 3, mmE5, and GME, on a MIEB-Lite benchmark (51 tasks) broken down by task categories.

<div align='center'><img src="https://cdn-uploads.huggingface.co/production/uploads/63108cc834c7d77420b0fd68/63WBsKh57HbNwwe3bZ-oZ.png" alt="mieb_lite" width="100%"/></div>

Performance and efficiency comparisons of different training strategies using 3B and 7B variants of Qwen2.5-VL backbones.

<div align='center'><img src="https://github.com/LCO-Embedding/LCO-Embedding/raw/main/assets/lora_ablation.png" alt="lora_ablation" width="100%"/></div>

Scaling relationship between generation benchmark performance (X-axis) and representation benchmark performance after language-centric contrastive learning (Y-axis).

<div align='center'><img src="https://github.com/LCO-Embedding/LCO-Embedding/raw/main/assets/scaling.png" alt="scaling_law" width="100%"/></div>

## Citation

If you find LCO-Embedding useful for your research and applications, please cite using this BibTeX:

```bibtex
@misc{xiao2025scaling,
    title={Scaling Language-Centric Omnimodal Representation Learning}, 
    author={Chenghao Xiao and Hou Pong Chan and Hao Zhang and Weiwen Xu and Mahani Aljunied and Yu Rong},
    year={2025},
    eprint={2510.11693},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    url={https://arxiv.org/abs/2510.11693}, 
}
```