Research Archive
Collection
Historical artifacts from SBERT-WK, RSE, InstructDS, and earlier research. No longer maintained; retained for reproducibility. โข 32 items โข Updated
How to use binwang/RSE-BERT-large-10-relations with Transformers:
# Load model directly
from transformers import AutoTokenizer, BertForRSE
tokenizer = AutoTokenizer.from_pretrained("binwang/RSE-BERT-large-10-relations")
model = BertForRSE.from_pretrained("binwang/RSE-BERT-large-10-relations", device_map="auto")No longer actively maintained. Retained for reproducibility of the original work.
This repository contains one variant of the RSE research models. The repository name identifies the backbone (BERT or RoBERTa), model size (base or large), and experiment variant (STS, USEB, Transfer, or 10-relations). Refer to the original description below for this checkpoint's configuration.
The RSE-BERT-large-10-rel is trained with 10 relations including:
The BERT-large-uncased model is used as initialization.
It can be used to infer all ten different relations.