README.md

---

language: 
  - en
tags:
- coreference-resolution
- xcore
- long-document
- cross-document
- maverick
license:   
- cc-by-nc-sa-4.0
datasets:
- ECB
metrics:
- CoNLL
task_categories:
- coreference-resolution
model-index:
- name: sapienzanlp/xcore-ecb
  results:
  - task:
      type: coreference-resolution
      name: coreference-resolution
    dataset:
      name: ECB
      type: coreference
    metrics:
    - name: Avg. F1
      type: CoNLL
      value: 42.4

---
# xCoRe ECB+
Official weights for *xcore*, pretrained on LitBank and trained on ECB+, based on DeBERTa-large.
This model achieves 42.4 Avg CoNLL-F1 on ECB+.

Other available models at [SapienzaNLP huggingface hub](https://huggingface.co/collections/sapienzanlp/xcore-models):

|            hf_model_name            | dataset | Score | Mode |
|:-----------------------------------:|:----------------:|:-----:|:----------:|
|    ["sapienzanlp/xcore-litbank"](https://huggingface.co/sapienzanlp/xcore-litbank)    |     [LitBank](https://aclanthology.org/2020.lrec-1.6/)    |  78.2 |    Long-Document  (Book Splits)|
|      ["sapienzanlp/xcore-ecb"](https://huggingface.co/sapienzanlp/xcore-ecb)      |       [ECB+](https://aclanthology.org/L14-1646/)      |  42.4 |     Cross-Document  (News)  |
|      ["sapienzanlp/xcore-scico"](https://huggingface.co/sapienzanlp/xcore-scico)      |       [SciCo](https://arxiv.org/abs/2104.08809)      |  31.0 |     Cross-Document (Scientific)    |
<!-- |    ["sapienzanlp/xcore-litbank"](https://huggingface.co/sapienzanlp/xcore-all)    |     LitBank, ECB+, BookCoref_silver, SciCo, OntoNotes, PreCo   |  - |   All datasets |-->
<!-- |     ["sapienzanlp/xcore-bookcoref"](https://huggingface.co/sapienzanlp/xcore-bookcoref)     |      [BookCoref](https://huggingface.co/datasets/sapienzanlp/bookcoref)     |  62.3 |     Single Document  (Full Books)  |-->


### Results on ECB+
<img src="https://cdn-uploads.huggingface.co/production/uploads/65e9ccd84ce78d665a50f78b/3c07se8JF5OCkXn0xxAnJ.png" alt="drawing" width="95%"/>



## xCoRe: Cross Context Coreference Resolution Defying recent trends
[![Conference](https://img.shields.io/badge/EMNLP%202025%20Paper-red)]()
[![License: CC BY-NC 4.0](https://img.shields.io/badge/License-CC%20BY--NC%204.0-green.svg)](https://creativecommons.org/licenses/by-nc/4.0/)
[![Pip Package](https://img.shields.io/badge/🐍%20Python%20package-blue)](https://pypi.org/project/xcore-coref/)
[![git](https://img.shields.io/badge/Git%20Repo%20-yellow.svg)](https://github.com/SapienzaNLP/xcore)
### Citation
This work has been published at [EMNLP 2025 main conference](https://aclanthology.org/2025.emnlp-main.1737/). If you use any part, please consider citing our paper as follows:
```bibtex
@inproceedings{martinelli-etal-2025-xcore,
    title = "x{C}o{R}e: Cross-context Coreference Resolution",
    author = "Martinelli, Giuliano  and
      Gatti, Bruno  and
      Navigli, Roberto",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2025.emnlp-main.1737/",
    pages = "34252--34266"
}
```