relbert/roberta-large-semeval2012-mask-prompt-e-nce
RelBERT fine-tuned from roberta-large on
relbert/semeval2012_relational_similarity.
Fine-tuning is done via RelBERT library (see the repository for more detail).
It achieves the following results on the relation understanding tasks:
- Analogy Question (dataset, full result):
- Accuracy on SAT (full): 0.6122994652406417
- Accuracy on SAT: 0.6172106824925816
- Accuracy on BATS: 0.7787659811006115
- Accuracy on U2: 0.6096491228070176
- Accuracy on U4: 0.6203703703703703
- Accuracy on Google: 0.924
- Lexical Relation Classification (dataset, full result):
- Micro F1 score on BLESS: 0.9328009642910954
- Micro F1 score on CogALexV: 0.8734741784037559
- Micro F1 score on EVALution: 0.7004333694474539
- Micro F1 score on K&H+N: 0.9617444529456771
- Micro F1 score on ROOT09: 0.9081792541523034
- Relation Mapping (dataset, full result):
- Accuracy on Relation Mapping: 0.8230555555555555
Usage
This model can be used through the relbert library. Install the library via pip
pip install relbert
and activate model as below.
from relbert import RelBERT
model = RelBERT("relbert/roberta-large-semeval2012-mask-prompt-e-nce")
vector = model.get_embedding(['Tokyo', 'Japan']) # shape of (1024, )
Training hyperparameters
The following hyperparameters were used during training:
- model: roberta-large
- max_length: 64
- mode: mask
- data: relbert/semeval2012_relational_similarity
- template_mode: manual
- template: I wasn’t aware of this relationship, but I just read in the encyclopedia that is ’s
- loss_function: nce_logout
- temperature_nce_constant: 0.05
- temperature_nce_rank: {'min': 0.01, 'max': 0.05, 'type': 'linear'}
- epoch: 23
- batch: 128
- lr: 5e-06
- lr_decay: False
- lr_warmup: 1
- weight_decay: 0
- random_seed: 0
- exclude_relation: None
- n_sample: 640
- gradient_accumulation: 8
The full configuration can be found at fine-tuning parameter file.
Reference
If you use any resource from RelBERT, please consider to cite our paper.
@inproceedings{ushio-etal-2021-distilling-relation-embeddings,
title = "{D}istilling {R}elation {E}mbeddings from {P}re-trained {L}anguage {M}odels",
author = "Ushio, Asahi and
Schockaert, Steven and
Camacho-Collados, Jose",
booktitle = "EMNLP 2021",
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
}