---
title: specter
canonical_url: "https://www.modelscope.cn/models/LLM-Research/specter"
md_url: "https://www.modelscope.cn/models/LLM-Research/specter.md"
repository: LLM-Research/specter
last_updated: 2024-11-21
license: apache-2.0
pipeline_tag: feature-extraction
tasks:
  - feature-extraction
model_type:
  - bert
architectures:
  - BertModel
library_name:
  - tensorflow
  - pytorch
frameworks:
  - pytorch
language:
  - en
inference_backends:
  - "deploy_task emb"
  - "sglang 0.5.2"
  - "vllm 0.9.2"
downloads: 2141
stars: 0
---

# specter

> specter - LLM-Research 在 ModelScope 开源的模型。SPECTER is a pre-trained language model to generate document-level embedding of documents. It is pre-trained on a powerful signal of document-level relatedness: the citation graph. Unlike existing pretrained language…

LLM-Research/specter 是 ModelScope 魔搭社区上的feature-extraction模型，采用 apache-2.0 许可，可用 deploy_task emb、sglang 0.5.2、vllm 0.9.2 部署。

- **Repository**: LLM-Research/specter
- **License**: apache-2.0
- **Tasks**: feature-extraction
- **Inference backends**: deploy_task emb, sglang 0.5.2, vllm 0.9.2
- **Downloads**: 2141
- **Stars**: 0
- **Last updated**: 2024-11-21

Source: https://www.modelscope.cn/models/LLM-Research/specter

---

## SPECTER

SPECTER is a pre-trained language model to generate document-level embedding of documents. It is pre-trained on a powerful signal of document-level relatedness: the citation graph. Unlike existing pretrained language models, SPECTER can be easily applied to downstream applications without task-specific fine-tuning. 

If you're coming here because you want to embed papers, SPECTER has now been superceded by [SPECTER2](https://huggingface.co/allenai/specter2_proximity). Use that instead.

Paper: [SPECTER: Document-level Representation Learning using Citation-informed Transformers](https://arxiv.org/pdf/2004.07180.pdf)

Original Repo: [Github](https://github.com/allenai/specter)

Evaluation Benchmark: [SciDocs](https://github.com/allenai/scidocs)

Authors: *Arman Cohan, Sergey Feldman, Iz Beltagy, Doug Downey, Daniel S. Weld*