scientific-multilingual-transfer
Collection
This Collection contains the Models from the paper [TBA Link] • 13 items • Updated
DE-Base-CP
German monolingual base model continued on the SciLaD target-language split as a 15k-step control baseline.
Model Details
This is a monolingual continued-pretraining control checkpoint reported in the paper table. It is provided for reproducibility of the baseline comparison.
- Paper name:
DE-Base-CP - Model role:
baseline-control - Source/base model: GermanT5/t5-efficient-gc4-german-base-nl36
- Code and pipeline: GitHub repository
- Architecture: T5 encoder-decoder
- SciLaD dataset: scilons/SciLaD-all-text-v1
- Evaluation benchmark: Global-MMLU
Evaluation
Zero-shot Global-MMLU accuracy reported by the paper aggregation:
| Metric | Accuracy |
|---|---|
| Average | 22.95 |
| STEM | 21.25 |
| Humanities | 24.21 |
| Social Sciences | 21.71 |
| Other | 23.98 |
Limitations
The model is evaluated primarily with zero-shot Global-MMLU. Downstream task-specific evaluation is recommended before deployment in specialized scientific workflows.
Citation
- Title: Transferring Scientific English Pre-Trained Language Models to Multiple Languages Using Cross-Lingual Transfer
- Authors: Nikolas Rauscher, Fabio Barth, Georg Rehm
- Venue: LREC-COLING 2026, citation details TBA after publication
- Downloads last month
- 13
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for rausch/de-t5-base-sci-cp-15k
Base model
GermanT5/t5-efficient-gc4-german-base-nl36