Update README.md
Browse files
README.md
CHANGED
|
@@ -1,6 +1,18 @@
|
|
| 1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
base_model:
|
| 3 |
- Qwen/Qwen3-4B-Instruct-2507
|
|
|
|
|
|
|
| 4 |
---
|
| 5 |
|
| 6 |
# Struct-SQL-8B: Knowledge Distillation with Structured Chain-of-Thought
|
|
@@ -11,7 +23,7 @@ Unlike standard distillation methods that rely on unstructured Chain-of-Thought
|
|
| 11 |
|
| 12 |
📄 **Paper:** [Knowledge Distillation with Structured Chain-of-Thought for Text-to-SQL](https://arxiv.org/abs/2512.17053)
|
| 13 |
*(Accepted at Canadian AI Conference 2026)*
|
| 14 |
-
|
| 15 |
|
| 16 |
## Performance
|
| 17 |
|
|
@@ -90,4 +102,4 @@ The model is not optimized for direct deployment in production database systems
|
|
| 90 |
booktitle={Proceedings of the 39th Canadian Conference on Artificial Intelligence},
|
| 91 |
year={2026},
|
| 92 |
note={To appear}
|
| 93 |
-
}
|
|
|
|
| 1 |
---
|
| 2 |
+
license: cc-by-4.0
|
| 3 |
+
library_name: transformers
|
| 4 |
+
tags:
|
| 5 |
+
- text-to-sql
|
| 6 |
+
- code
|
| 7 |
+
- qwen3
|
| 8 |
+
- knowledge-distillation
|
| 9 |
+
datasets:
|
| 10 |
+
- bird-bench/bird # Links to the official BIRD dataset
|
| 11 |
+
- craterlabs/struct-sql-synth-1k # REPLACE this with your actual dataset ID
|
| 12 |
base_model:
|
| 13 |
- Qwen/Qwen3-4B-Instruct-2507
|
| 14 |
+
language:
|
| 15 |
+
- en
|
| 16 |
---
|
| 17 |
|
| 18 |
# Struct-SQL-8B: Knowledge Distillation with Structured Chain-of-Thought
|
|
|
|
| 23 |
|
| 24 |
📄 **Paper:** [Knowledge Distillation with Structured Chain-of-Thought for Text-to-SQL](https://arxiv.org/abs/2512.17053)
|
| 25 |
*(Accepted at Canadian AI Conference 2026)*
|
| 26 |
+
|
| 27 |
|
| 28 |
## Performance
|
| 29 |
|
|
|
|
| 102 |
booktitle={Proceedings of the 39th Canadian Conference on Artificial Intelligence},
|
| 103 |
year={2026},
|
| 104 |
note={To appear}
|
| 105 |
+
}
|