MultiIndic WikiBio Separate Scripts
This biography generation release fine-tunes IndicBARTSS on the corresponding IndicNLG dataset. It supports 9 Indian languages.
The task fine-tune is separate from the pretrained model
Its separate-script design follows the base model. No Devanagari round trip is needed. The release is intended for biography generation; it is not a general conversation model.MultiIndic WikiBio Separate Scripts official model card
Review the generated text against its input
Practical guidance. Use a source record with names, dates and occupations. Check generated statements against that record; fluent prose is not evidence for a new biographical fact.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | MultiIndicWikiBioSS |
| Input scripts | Native scripts are retained.MultiIndic WikiBio Separate Scripts official model card |
| Training set | 34,653 examplesMultiIndic WikiBio Separate Scripts official model card |
| License | MIT in the model card.MultiIndic WikiBio Separate Scripts official model card |
Related models
- IndicBART Separate Scripts
AI4Bharat · IndicBART
IndicBART generation variant that retains each Indian language's own script.
Launch date not confirmedDownloadable - MultiIndic WikiBio Unified Script
AI4Bharat · IndicNLG
Unified-script IndicBART adaptation for biography generation in nine Indian languages.
Launch date not confirmedDownloadable