---
license: apache-2.0
---

# Beam Retrieval: General End-to-End Retrieval for Multi-Hop Question Answering (Zhang et all 2023)

Unofficial mirror of [Beam Retriever](https://github.com/canghongjian/beam_retriever)

This is the finetuned **encoder only** [DebertaV3Large](https://huggingface.co/microsoft/deberta-v3-large) of the Beam Retriever model which can be used for maximum inner product search.

## Usage

```python
from transformers import DebertaV2Model

finetuned_encoder = DebertaV2Model.from_pretrained('scholarly-shadows-syndicate/beam_retriever_unofficial_encoder_only')
```

## Citations

```bibtex
@article{Zhang2023BeamRG,
  title={Beam Retrieval: General End-to-End Retrieval for Multi-Hop Question Answering},
  author={Jiahao Zhang and H. Zhang and Dongmei Zhang and Yong Liu and Sheng Huang},
  journal={ArXiv},
  year={2023},
  volume={abs/2308.08973},
  url={https://api.semanticscholar.org/CorpusID:261030563}
}
```

```bibtex
@article{He2020DeBERTaDB,
  title={DeBERTa: Decoding-enhanced BERT with Disentangled Attention},
  author={Pengcheng He and Xiaodong Liu and Jianfeng Gao and Weizhu Chen},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.03654},
  url={https://api.semanticscholar.org/CorpusID:219531210}
}
```