Laya Task Area v2
Area-classification specialist fine-tuned from convaiinnovations/laya. It routes task text to one of three areas: home, work, or personal.
This checkpoint was trained on the area configuration of chand1012/laya-task-routing-synthetic-v2.
Evaluation
On the held-out synthetic v2 test split (181 records):
| Metric | Result |
|---|---|
| Accuracy | 96.69% |
| Macro-F1 | 96.67% |
| Record exact match | 96.69% |
| ECE | 0.69% |
See evaluation_report.json for the confusion matrix, class-level F1, calibration, and latency results. These are synthetic-set results and should not be interpreted as production performance.
Usage
import laya
agent = laya.load("chand1012/laya-task-area-v2")
result = agent.predict(state, questions)
Limitations
The training and evaluation data are synthetic and English-only. The model always chooses among the three configured areas; applications should add abstention or human review where ambiguity and out-of-taxonomy inputs matter.
License
Apache License 2.0. This is a derivative of the Apache-2.0-licensed Laya base model.