Update README with corrected card and Denash org references
Browse files
README.md
CHANGED
|
@@ -2,7 +2,7 @@
|
|
| 2 |
|
| 3 |
> HuggingFace-ready model card. This documents the custom Stage-2 classifier
|
| 4 |
> trained by `src/training/` and is the artifact to publish at
|
| 5 |
-
> `huggingface.co/
|
| 6 |
|
| 7 |
---
|
| 8 |
|
|
@@ -10,7 +10,7 @@
|
|
| 10 |
|
| 11 |
### Model Description
|
| 12 |
|
| 13 |
-
- **Model ID**: `
|
| 14 |
- **Model type**: Fine-tuned transformer for binary sequence classification
|
| 15 |
- **Base model**: [`microsoft/codebert-base`](https://huggingface.co/microsoft/codebert-base) (RoBERTa-based, ~125 M parameters)
|
| 16 |
- **Language**: Python (code)
|
|
@@ -82,7 +82,7 @@ pipeline. Its intended use case is:
|
|
| 82 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 83 |
import torch
|
| 84 |
|
| 85 |
-
model_id = "
|
| 86 |
|
| 87 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 88 |
model = AutoModelForSequenceClassification.from_pretrained(model_id)
|
|
@@ -443,7 +443,7 @@ pip install transformers torch
|
|
| 443 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 444 |
import torch
|
| 445 |
|
| 446 |
-
model_id = "
|
| 447 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 448 |
model = AutoModelForSequenceClassification.from_pretrained(model_id)
|
| 449 |
model.eval()
|
|
@@ -507,10 +507,20 @@ CEVuD Authors
|
|
| 507 |
title={CEVuD: Cost-Effective Vulnerability Detection via Gated Static-Neural Reasoning},
|
| 508 |
author={CEVuD Authors},
|
| 509 |
year={2026},
|
| 510 |
-
note={Model:
|
| 511 |
}
|
| 512 |
```
|
| 513 |
|
| 514 |
## Model Card Contact
|
| 515 |
|
| 516 |
Open an issue on the CEVuD GitHub repository.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2 |
|
| 3 |
> HuggingFace-ready model card. This documents the custom Stage-2 classifier
|
| 4 |
> trained by `src/training/` and is the artifact to publish at
|
| 5 |
+
> `huggingface.co/Denash/codebert-vuln-classifier`.
|
| 6 |
|
| 7 |
---
|
| 8 |
|
|
|
|
| 10 |
|
| 11 |
### Model Description
|
| 12 |
|
| 13 |
+
- **Model ID**: `Denash/codebert-vuln-classifier`
|
| 14 |
- **Model type**: Fine-tuned transformer for binary sequence classification
|
| 15 |
- **Base model**: [`microsoft/codebert-base`](https://huggingface.co/microsoft/codebert-base) (RoBERTa-based, ~125 M parameters)
|
| 16 |
- **Language**: Python (code)
|
|
|
|
| 82 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 83 |
import torch
|
| 84 |
|
| 85 |
+
model_id = "Denash/codebert-vuln-classifier"
|
| 86 |
|
| 87 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 88 |
model = AutoModelForSequenceClassification.from_pretrained(model_id)
|
|
|
|
| 443 |
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 444 |
import torch
|
| 445 |
|
| 446 |
+
model_id = "Denash/codebert-vuln-classifier"
|
| 447 |
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 448 |
model = AutoModelForSequenceClassification.from_pretrained(model_id)
|
| 449 |
model.eval()
|
|
|
|
| 507 |
title={CEVuD: Cost-Effective Vulnerability Detection via Gated Static-Neural Reasoning},
|
| 508 |
author={CEVuD Authors},
|
| 509 |
year={2026},
|
| 510 |
+
note={Model: Denash/codebert-vuln-classifier; Training Dataset: Denash/cevud-training-dataset; Pipeline Dataset: Denash/cevud-pipeline-dataset}
|
| 511 |
}
|
| 512 |
```
|
| 513 |
|
| 514 |
## Model Card Contact
|
| 515 |
|
| 516 |
Open an issue on the CEVuD GitHub repository.
|
| 517 |
+
|
| 518 |
+
## Related Resources
|
| 519 |
+
|
| 520 |
+
| Resource | Link |
|
| 521 |
+
|----------|------|
|
| 522 |
+
| **Training Dataset (CVEfixes)** | [`Denash/cevud-training-dataset`](https://huggingface.co/datasets/Denash/cevud-training-dataset) |
|
| 523 |
+
| **Pipeline Dataset (VUDENC)** | [`Denash/cevud-pipeline-dataset`](https://huggingface.co/datasets/Denash/cevud-pipeline-dataset) |
|
| 524 |
+
| **Source Dataset (CVEfixes)** | [`hitoshura25/cvefixes`](https://huggingface.co/datasets/hitoshura25/cvefixes) |
|
| 525 |
+
| **Source Dataset (VUDENC)** | [`DetectVul/Vudenc`](https://huggingface.co/datasets/DetectVul/Vudenc) |
|
| 526 |
+
| **CEVuD GitHub** | https://github.com/Denash/CEVuD |
|