{"id":38676306,"url":"https://github.com/wglab/bioformer","last_synced_at":"2026-01-17T10:01:06.543Z","repository":{"id":48216470,"uuid":"359677815","full_name":"WGLab/Bioformer","owner":"WGLab","description":"Bioformer: an efficient BERT model for biomedical text mining","archived":false,"fork":false,"pushed_at":"2023-02-07T04:09:44.000Z","size":30,"stargazers_count":55,"open_issues_count":0,"forks_count":6,"subscribers_count":6,"default_branch":"main","last_synced_at":"2025-12-09T09:04:29.478Z","etag":null,"topics":["bert-model","biomedical-nlp","natural-language-processing","natural-language-understanding","nlp"],"latest_commit_sha":null,"homepage":"","language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/WGLab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2021-04-20T03:56:49.000Z","updated_at":"2025-05-20T03:19:51.000Z","dependencies_parsed_at":"2022-09-15T00:20:16.369Z","dependency_job_id":null,"html_url":"https://github.com/WGLab/Bioformer","commit_stats":null,"previous_names":[],"tags_count":3,"template":false,"template_full_name":null,"purl":"pkg:github/WGLab/Bioformer","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/WGLab%2FBioformer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/WGLab%2FBioformer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/WGLab%2FBioformer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/WGLab%2FBioformer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/WGLab","download_url":"https://codeload.github.com/WGLab/Bioformer/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/WGLab%2FBioformer/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28505570,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-17T06:57:29.758Z","status":"ssl_error","status_checked_at":"2026-01-17T06:56:03.931Z","response_time":85,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["bert-model","biomedical-nlp","natural-language-processing","natural-language-understanding","nlp"],"created_at":"2026-01-17T10:00:43.893Z","updated_at":"2026-01-17T10:01:06.522Z","avatar_url":"https://github.com/WGLab.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Bioformer: an efficient BERT model for biomedical text mining\nBioformer is a lightweight BERT model pretrained from biomedical Literature. We pretrained two Bioformer models, `Bioformer-8L` and `Bioformer-16L`. Both models were pretrained on all PubMed abstracts (as of Jan 2021) and 1 million subsampled PubMed Central full-text articles. We used the original implementation of [BERT](https://github.com/google-research/bert) to train the model. Bioformer models have the following features:\n \n - **Accurate**. Bioformer achieves comparable or even better performance than BioBERT/PubMedBERT on downstream NLP tasks. A detailed evaluation is [here](https://arxiv.org/abs/2302.01588).\n - **Smaller model size**. `Bioformer-8L` and `Bioformer-16L` reduced the model size by 60% compared with `BERT-Base`/`BioBERT-Base`/`PubMedBERT`.\n - **Fast and memory efficient**. `Bioformer-8L` is 3X as fast as `PubMedBERT`, and `Bioformer-16L` is 2X as fast as `PubMedBERT`.\n - **Biomedical vocabulary**. Bioformer uses a biomedical vocabulary of 32768 tokens, which was trained from PubMed abstracts and PubMed Central full-text articles. Bioformer is able to encode some special unicode symbols that are not in the original BERT vocabulary. \n\n\n## Download \n\n#### Pytorch checkpoint\n\nPretrained model weights of `Bioformer-8L` and `Bioformer-16L` are available on HuggingFace ([Bioformer-8L](https://huggingface.co/bioformers/bioformer-8L), and [Bioformer-16L](https://huggingface.co/bioformers/bioformer-16L))\n\nYou can easily use Bioformer with the [transformers](https://github.com/huggingface/transformers) library. \n\n\n## Acknowledgment\n\nPretraining of Bioformer is partly supported by the Google TPU Research Cloud (TRC) program.\n\n## Citation\n\nFang L, Chen Q, Wei C-H, Lu Z, Wang K: Bioformer: an efficient transformer language model for biomedical text mining. arXiv preprint arXiv:2302.01588 (2023). DOI: https://doi.org/10.48550/arXiv.2302.01588\n\n```\n@ARTICLE{fangli2023bioformer,\n       author = {{Fang}, Li and {Chen}, Qingyu and {Wei}, Chih-Hsuan and {Lu}, Zhiyong and {Wang}, Kai},\n        title = \"{Bioformer: an efficient transformer language model for biomedical text mining}\",\n      journal = {arXiv preprint arXiv:2302.01588},\n         year = {2023}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwglab%2Fbioformer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fwglab%2Fbioformer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwglab%2Fbioformer/lists"}