Datasets:
Update README.md
Browse files
README.md
CHANGED
|
@@ -12,6 +12,8 @@ tags:
|
|
| 12 |
- instruction-tuning
|
| 13 |
- benchmark
|
| 14 |
pretty_name: TransitLM
|
|
|
|
|
|
|
| 15 |
---
|
| 16 |
|
| 17 |
# TransitLM: Dataset Release & Evaluation Protocol
|
|
@@ -172,4 +174,4 @@ These properties make it infeasible to associate records with the same individua
|
|
| 172 |
| Reproducibility | ✅ Full training pipeline reproducible | ✅ Standardized community test set |
|
| 173 |
| Training utility | ✅ < 0.1% spatial error, topology preserved | Not used for training |
|
| 174 |
|
| 175 |
-
This dual-release design maximizes research utility through reproducible training and fair evaluation, while minimizing privacy exposure by applying perturbation only where data scale creates risk.
|
|
|
|
| 12 |
- instruction-tuning
|
| 13 |
- benchmark
|
| 14 |
pretty_name: TransitLM
|
| 15 |
+
size_categories:
|
| 16 |
+
- 10B<n<100B
|
| 17 |
---
|
| 18 |
|
| 19 |
# TransitLM: Dataset Release & Evaluation Protocol
|
|
|
|
| 174 |
| Reproducibility | ✅ Full training pipeline reproducible | ✅ Standardized community test set |
|
| 175 |
| Training utility | ✅ < 0.1% spatial error, topology preserved | Not used for training |
|
| 176 |
|
| 177 |
+
This dual-release design maximizes research utility through reproducible training and fair evaluation, while minimizing privacy exposure by applying perturbation only where data scale creates risk.
|