Model Card for impresso-project/nel-mgenre-multilingual-light

The Impresso multilingual named entity linking (NEL) model is based on mGENRE (multilingual Generative ENtity REtrieval) proposed by De Cao et al, a sequence-to-sequence architecture for entity disambiguation based on mBART. It uses constrained generation to output entity names mapped to Wikidata/QIDs.

This model was adapted for historical texts and fine-tuned on the HIPE-2022 dataset, which includes a variety of historical document types and languages.

Model Details

Model Description

Model Description

• Developed by: EPFL from the Impresso team. The project is an interdisciplinary project focused on historical media analysis across languages, time, and modalities. Funded by the Swiss National Science Foundation (CRSII5_173719, CRSII5_213585) and the Luxembourg National Research Fund (grant No. 17498891).
• Model type: mBART-based sequence-to-sequence model with constrained beam search for named entity linking
• Languages: Multilingual (100+ languages, optimized for French, German, and English)
• License: AGPL v3+
• Finetuned from: facebook/mgenre-wiki

Model Architecture

• Architecture: mBART-based seq2seq with constrained beam search

Training Details

Training Data

The model was trained on the following datasets:

Dataset alias	README	Document type	Languages	Suitable for	Project	License
ajmc	link	classical commentaries	de, fr, en	NERC-Coarse, NERC-Fine, EL	AjMC
hipe2020	link	historical newspapers	de, fr, en	NERC-Coarse, NERC-Fine, EL	CLEF-HIPE-2020
topres19th	link	historical newspapers	en	NERC-Coarse, EL	Living with Machines
newseye	link	historical newspapers	de, fi, fr, sv	NERC-Coarse, NERC-Fine, EL	NewsEye
sonar	link	historical newspapers	de	NERC-Coarse, EL	SoNAR

How to Use

from transformers import AutoTokenizer, pipeline

NEL_MODEL_NAME = "impresso-project/nel-mgenre-multilingual-light"
nel_tokenizer = AutoTokenizer.from_pretrained(NEL_MODEL_NAME)

nel_pipeline = pipeline("generic-nel", model=NEL_MODEL_NAME,
                        tokenizer=nel_tokenizer,
                        trust_remote_code=True,
                        device='cpu')

sentence = "Le 0ctobre 1894, [START] Dreyfvs [END] est arrêté à Paris, accusé d'espionnage pour l'Allemagne — un événement qui déch1ra la société fr4nçaise pendant des années."
print(nel_pipeline(sentence))

Output Format

[
  {'surface': 'Dreyfvs', 'wkd_pred': 'Alfred Dreyfus >> fr ',
  'type': 'UNK', 'confidence_nel': 100.0, 'lOffset': 24, 'rOffset': 33},
  {'surface': 'Dreyfvs', 'wkd_pred': 'Alfred Dreyfus >> fr', 'type': 'UNK',
  'confidence_nel': 41.0, 'lOffset': 24, 'rOffset': 33}, {'surface': 'Dreyfvs',
  'wkd_pred': 'Alfred Dreyfuss >> fr ', 'type': 'UNK', 'confidence_nel': 38.0,
  'lOffset': 24, 'rOffset': 33}, {'surface': 'Dreyfvs', 'wkd_pred': 'Alfred Dreyfus >> fr ',
  'type': 'UNK', 'confidence_nel': 26.0, 'lOffset': 24, 'rOffset': 33}, {'surface': 'Dreyfvs',
  'wkd_pred': 'Alfred Dreyfw >> fr ', 'type': 'UNK', 'confidence_nel': 24.0, 'lOffset': 24, 'rOffset': 33}]

The type of the entity is UNK because the model was not trained on the entity type. The confidence_nel score indicates the model's confidence in the prediction.

Use Cases

• Entity disambiguation in noisy OCR settings
• Linking historical names to modern Wikidata entities
• Assisting downstream event extraction and biography generation from historical archives

Limitations

• Sensitive to tokenisation and malformed spans
• Accuracy degrades on non-Wikidata entities or in highly ambiguous contexts
• Focused on historical entity mentions — performance may vary on modern texts

Environmental Impact

• Hardware: 1x A100 (80GB) for finetuning
• Training time: ~12 hours
• Estimated CO₂ Emissions: ~2.3 kg CO₂eq

Contact

• Website: https://impresso-project.ch