thanhdath commited on
Commit
3624525
·
verified ·
1 Parent(s): 9684f35

Replace README with minimal version (GitHub link + citation)

Browse files
Files changed (1) hide show
  1. README.md +13 -17
README.md CHANGED
@@ -1,17 +1,13 @@
1
- ---
2
- base_model:
3
- - griffith-bigdata/Qwen-2.5-Coder-3B-SQL-Writer
4
- ---
5
-
6
- # FINER-SQL-3B-BIRD
7
-
8
- Trained from [`griffith-bigdata/Qwen-2.5-Coder-3B-SQL-Writer`](https://huggingface.co/griffith-bigdata/Qwen-2.5-Coder-3B-SQL-Writer) using GRPO with two dense rewards from the FINER-SQL paper:
9
-
10
- 🧠 Memory Reward — aligns reasoning with verified traces
11
- ⚙️ Atomic Reward — measures operation-level SQL overlap
12
-
13
- ✅ 67.5% EX on BIRD when training only on BIRD train dataset; infer on a single 12-24 GB GPU
14
-
15
- 📄 See other models: https://huggingface.co/collections/griffith-bigdata/finer-sql
16
-
17
- 📄 Github: https://github.com/thanhdath/finer-sql/tree/main
 
1
+ 📄 GitHub: https://github.com/thanhdath/finer-sql
2
+
3
+ ## Citation
4
+
5
+ ```bibtex
6
+ @inproceedings{finersql,
7
+ author = {Thanh Dat Hoang and Thanh Trung Huynh and Matthias Weidlich and Thanh Tam Nguyen and Tong Chen and Hongzhi Yin and Quoc Viet Hung Nguyen},
8
+ title = {Boosting Small Language Models for Text-to-SQL with Fine-Grained Execution Feedback and Cost-Efficient Rewards},
9
+ booktitle = {ICDE},
10
+ publisher = {IEEE},
11
+ year = {2026},
12
+ }
13
+ ```