BERT masked-LM architecture
The captured configuration identifies BertForMaskedLM with model type bert and Transformers support.
Open Source Model Profile · jjzha
jobbert-base-cased
jobbert-base-cased is a 108M-parameter BERT-family fill-mask model from jjzha. According to the model card, it continues bert-base-cased on job postings for skill extraction.
- Publisher
- jjzha
- Task
- fill-mask
- Model type
- bert
- License
- Unknown
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
jobbert-base-cased is published by jjzha as a fill-mask model. The captured configuration identifies BertForMaskedLM with model type bert, and Safetensors metadata reports about 108M parameters. According to the model card, it is the JobBERT model continuously pretrained from bert-base-cased on about 3.2M job-posting sentences.
Recorded capabilities
108M job-posting adaptation
Safetensors metadata reports 108,341,316 parameters; according to the model card, the model continues from bert-base-cased on about 3.2M job-posting sentences.
Job-posting domain tagging
Hub tags record bert, fill-mask, JobBERT, job postings, and en with endpoints compatibility.
SkillSpan research provenance
According to the model card, the model comes from the SkillSpan hard and soft skill-extraction work published at NAACL 2022, with a full citation provided.
Use cases in the source record
- Fill-mask and skill-extraction experiments on English job postings consistent with the SkillSpan hard and soft skill focus.
- Domain-adapted language-modeling research building on the publisher-described job-posting continued pretraining.
Limitations and unknowns
- No license value was extracted from this record.
- No evaluation results were extracted in structured form.
- No context-window value was extracted from this record.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: jjzha/jobbert-base-cased
Captured: Unknown. Processed: 2026-09-07T19:34:48.192566+00:00.
JobBERT This is the JobBERT model from: Mike Zhang, Kristian Nørgaard Jensen, Sif Dam Sonniks, and Barbara Plank. SkillSpan: Hard and Soft Skill Extraction from Job Postings . Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. This model is continuously pre-trained from a bert-base-cased checkpoint on ~3.2M sentences from job postings. More information can be found in the paper. If you use this model, please cite the following paper: @inproceedings{zhang-etal-2022-skillspan, title = "{S}kill{S}pan: Hard and Soft Skill Extraction from {E}ngli…
F001F002F003F004F005F006F007F008F009F010F011F012F013