Laya fine-tuned for climate-vulnerability group detection (multi-label)
convaiinnovations/laya (421M, ModernBERT-large backbone) fine-tuned with the official RLCD recipe on GIZ/vulnerability_training_data_full (380 train rows x 17 binary vulnerability-group questions; 36 all-negative rows included as negatives).
Each vulnerability group is asked as one binary (noul) typed question; all 17 are answered in a single forward pass.
Evaluate with laya.load("peter2000/laya-vulnerability-groups") and agent.predict(state, questions).
Test-set metrics (held-out 95 rows, threshold 0.5)
| metric | value |
|---|---|
| macro-F1 | 0.6544 |
| micro-F1 | 0.6694 |
| ECE | 0.0263 |
| subset accuracy | 0.4737 |
Per-label F1:
| label | F1 |
|---|---|
| Agricultural communities | 0.9565 |
| Coastal communities | 0.5000 |
| Ethnic, racial or other minorities | 0.6000 |
| Fishery communities | 0.4000 |
| Informal sector workers | 1.0000 |
| Members of indigenous and local communities | 0.9333 |
| Migrants and displaced persons | 0.5714 |
| Older persons | 0.8889 |
| Other | 0.0000 |
| Persons living in poverty | 0.3333 |
| Persons with disabilities | 1.0000 |
| Persons with pre-existing health conditions | 0.8000 |
| Residents of drought-prone regions | 0.5714 |
| Rural populations | 0.8000 |
| Sexual minorities (LGBTQI+) | 0.3333 |
| Urban populations | 0.5714 |
| Women and other genders | 0.8649 |