Instructions to use MichaelErchi/CodeVerifier with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use MichaelErchi/CodeVerifier with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("MichaelErchi/CodeVerifier", device_map="auto") - Notebooks
- Google Colab
- Kaggle
docs: clarify verifier use cases
Browse files
README.md
CHANGED
|
@@ -22,7 +22,9 @@ tags:
|
|
| 22 |
|
| 23 |
## 🌐 Why CodeVerifier?
|
| 24 |
|
| 25 |
-
|
|
|
|
|
|
|
| 26 |
|
| 27 |
## ✨ Highlights
|
| 28 |
|
|
|
|
| 22 |
|
| 23 |
## 🌐 Why CodeVerifier?
|
| 24 |
|
| 25 |
+
Execution provides reliable feedback, but it often arrives too late for large candidate pools, online policy updates, and repository workflows. CodeVerifier supplies timely reward signals for reinforcement learning with verifiable rewards (RLVR) and candidate search, as well as rapid quality feedback for filtering generated training data before a full test run.
|
| 26 |
+
|
| 27 |
+
Rather than returning only a scalar score, CodeVerifier produces a verdict, a brief explanation, and supporting code regions for complete programs, intermediate edits, and repository code states. It learns these judgments from historical execution outcomes and successful repairs through evidence projection and RVPG.
|
| 28 |
|
| 29 |
## ✨ Highlights
|
| 30 |
|