Text Classification
Transformers
Safetensors
English
roberta
Generated from Trainer
classification
nlp
vulnerability
Eval Results (legacy)
text-embeddings-inference
Instructions to use CIRCL/vulnerability-severity-classification-roberta-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use CIRCL/vulnerability-severity-classification-roberta-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="CIRCL/vulnerability-severity-classification-roberta-base")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("CIRCL/vulnerability-severity-classification-roberta-base") model = AutoModelForSequenceClassification.from_pretrained("CIRCL/vulnerability-severity-classification-roberta-base", device_map="auto") - Inference
- Notebooks
- Google Colab
- Kaggle
End of training
Browse files- README.md +35 -70
- config.json +1 -1
- emissions.csv +1 -1
- model.safetensors +1 -1
README.md
CHANGED
|
@@ -1,69 +1,50 @@
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
-
license:
|
| 4 |
base_model: roberta-base
|
| 5 |
-
metrics:
|
| 6 |
-
- accuracy
|
| 7 |
tags:
|
| 8 |
- generated_from_trainer
|
| 9 |
-
|
| 10 |
-
-
|
| 11 |
-
- nlp
|
| 12 |
-
- vulnerability
|
| 13 |
model-index:
|
| 14 |
- name: vulnerability-severity-classification-roberta-base
|
| 15 |
results: []
|
| 16 |
-
datasets:
|
| 17 |
-
- CIRCL/vulnerability-scores
|
| 18 |
---
|
| 19 |
|
|
|
|
|
|
|
| 20 |
|
| 21 |
-
#
|
| 22 |
-
|
| 23 |
-
# Severity classification
|
| 24 |
-
|
| 25 |
-
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the dataset [CIRCL/vulnerability-scores](https://huggingface.co/datasets/CIRCL/vulnerability-scores).
|
| 26 |
-
|
| 27 |
-
The model was presented in the paper [VLAI: A RoBERTa-Based Model for Automated Vulnerability Severity Classification](https://huggingface.co/papers/2507.03607) [[arXiv](https://arxiv.org/abs/2507.03607)].
|
| 28 |
-
|
| 29 |
-
**Abstract:** VLAI is a transformer-based model that predicts software vulnerability severity levels directly from text descriptions. Built on RoBERTa, VLAI is fine-tuned on over 600,000 real-world vulnerabilities and achieves over 82% accuracy in predicting severity categories, enabling faster and more consistent triage ahead of manual CVSS scoring. The model and dataset are open-source and integrated into the Vulnerability-Lookup service.
|
| 30 |
-
|
| 31 |
-
You can read [this page](https://www.vulnerability-lookup.org/user-manual/ai/) for more information.
|
| 32 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
|
| 34 |
## Model description
|
| 35 |
|
| 36 |
-
|
| 37 |
-
|
| 38 |
-
## How to get started with the model
|
| 39 |
-
|
| 40 |
-
```python
|
| 41 |
-
from transformers import AutoModelForSequenceClassification, AutoTokenizer
|
| 42 |
-
import torch
|
| 43 |
|
| 44 |
-
|
| 45 |
|
| 46 |
-
|
| 47 |
-
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
| 48 |
-
model = AutoModelForSequenceClassification.from_pretrained(model_name)
|
| 49 |
-
model.eval()
|
| 50 |
|
| 51 |
-
|
| 52 |
|
| 53 |
-
|
| 54 |
-
that could severely harm the host system. This could significantly affect the confidentiality, integrity, and availability of the targeted system."
|
| 55 |
-
inputs = tokenizer(test_description, return_tensors="pt", truncation=True, padding=True)
|
| 56 |
-
|
| 57 |
-
# Run inference
|
| 58 |
-
with torch.no_grad():
|
| 59 |
-
outputs = model(**inputs)
|
| 60 |
-
predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
|
| 61 |
-
|
| 62 |
-
# Print results
|
| 63 |
-
print("Predictions:", predictions)
|
| 64 |
-
predicted_class = torch.argmax(predictions, dim=-1).item()
|
| 65 |
-
print("Predicted severity:", labels[predicted_class])
|
| 66 |
-
```
|
| 67 |
|
| 68 |
## Training procedure
|
| 69 |
|
|
@@ -82,32 +63,16 @@ The following hyperparameters were used during training:
|
|
| 82 |
|
| 83 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Low Precision | Low Recall | Low F1 | Medium Precision | Medium Recall | Medium F1 | High Precision | High Recall | High F1 | Critical Precision | Critical Recall | Critical F1 |
|
| 84 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:-------------:|:----------:|:------:|:----------------:|:-------------:|:---------:|:--------------:|:-----------:|:-------:|:------------------:|:---------------:|:-----------:|
|
| 85 |
-
| 2.
|
| 86 |
-
| 2.
|
| 87 |
-
|
|
| 88 |
-
| 1.
|
| 89 |
-
| 1.
|
| 90 |
|
| 91 |
-
It achieves the following results on the evaluation set:
|
| 92 |
-
- Loss: 2.0802
|
| 93 |
-
- Accuracy: 0.8101
|
| 94 |
-
- F1 Macro: 0.7378
|
| 95 |
-
- Low Precision: 0.6473
|
| 96 |
-
- Low Recall: 0.4729
|
| 97 |
-
- Low F1: 0.5465
|
| 98 |
-
- Medium Precision: 0.8264
|
| 99 |
-
- Medium Recall: 0.8726
|
| 100 |
-
- Medium F1: 0.8489
|
| 101 |
-
- High Precision: 0.8190
|
| 102 |
-
- High Recall: 0.7991
|
| 103 |
-
- High F1: 0.8089
|
| 104 |
-
- Critical Precision: 0.7599
|
| 105 |
-
- Critical Recall: 0.7339
|
| 106 |
-
- Critical F1: 0.7467
|
| 107 |
|
| 108 |
### Framework versions
|
| 109 |
|
| 110 |
-
- Transformers 5.
|
| 111 |
-
- Pytorch 2.14.
|
| 112 |
- Datasets 4.8.5
|
| 113 |
- Tokenizers 0.23.2
|
|
|
|
| 1 |
---
|
| 2 |
library_name: transformers
|
| 3 |
+
license: mit
|
| 4 |
base_model: roberta-base
|
|
|
|
|
|
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
|
|
|
|
|
|
| 9 |
model-index:
|
| 10 |
- name: vulnerability-severity-classification-roberta-base
|
| 11 |
results: []
|
|
|
|
|
|
|
| 12 |
---
|
| 13 |
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
|
| 17 |
+
# vulnerability-severity-classification-roberta-base
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.0906
|
| 22 |
+
- Accuracy: 0.8109
|
| 23 |
+
- F1 Macro: 0.7383
|
| 24 |
+
- Low Precision: 0.6926
|
| 25 |
+
- Low Recall: 0.4428
|
| 26 |
+
- Low F1: 0.5402
|
| 27 |
+
- Medium Precision: 0.8274
|
| 28 |
+
- Medium Recall: 0.8662
|
| 29 |
+
- Medium F1: 0.8464
|
| 30 |
+
- High Precision: 0.8130
|
| 31 |
+
- High Recall: 0.8124
|
| 32 |
+
- High F1: 0.8127
|
| 33 |
+
- Critical Precision: 0.7677
|
| 34 |
+
- Critical Recall: 0.7406
|
| 35 |
+
- Critical F1: 0.7539
|
| 36 |
|
| 37 |
## Model description
|
| 38 |
|
| 39 |
+
More information needed
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 40 |
|
| 41 |
+
## Intended uses & limitations
|
| 42 |
|
| 43 |
+
More information needed
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
+
## Training and evaluation data
|
| 46 |
|
| 47 |
+
More information needed
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 48 |
|
| 49 |
## Training procedure
|
| 50 |
|
|
|
|
| 63 |
|
| 64 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | Low Precision | Low Recall | Low F1 | Medium Precision | Medium Recall | Medium F1 | High Precision | High Recall | High F1 | Critical Precision | Critical Recall | Critical F1 |
|
| 65 |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:-------------:|:----------:|:------:|:----------------:|:-------------:|:---------:|:--------------:|:-----------:|:-------:|:------------------:|:---------------:|:-----------:|
|
| 66 |
+
| 2.3598 | 1.0 | 19363 | 2.6454 | 0.7298 | 0.6067 | 0.6491 | 0.1836 | 0.2862 | 0.7618 | 0.8161 | 0.7880 | 0.7188 | 0.7317 | 0.7252 | 0.6449 | 0.6110 | 0.6275 |
|
| 67 |
+
| 2.3058 | 2.0 | 38726 | 2.3857 | 0.7570 | 0.6550 | 0.6429 | 0.2886 | 0.3984 | 0.7937 | 0.8234 | 0.8083 | 0.7285 | 0.7860 | 0.7562 | 0.7298 | 0.5979 | 0.6573 |
|
| 68 |
+
| 2.0088 | 3.0 | 58089 | 2.2418 | 0.7775 | 0.6871 | 0.6675 | 0.3581 | 0.4661 | 0.8082 | 0.8369 | 0.8223 | 0.7501 | 0.8133 | 0.7804 | 0.7771 | 0.6041 | 0.6798 |
|
| 69 |
+
| 1.5341 | 4.0 | 77452 | 2.1168 | 0.8009 | 0.7309 | 0.6294 | 0.4803 | 0.5448 | 0.8208 | 0.8561 | 0.8381 | 0.7964 | 0.8080 | 0.8021 | 0.7860 | 0.6963 | 0.7384 |
|
| 70 |
+
| 1.4963 | 5.0 | 96815 | 2.0906 | 0.8109 | 0.7383 | 0.6926 | 0.4428 | 0.5402 | 0.8274 | 0.8662 | 0.8464 | 0.8130 | 0.8124 | 0.8127 | 0.7677 | 0.7406 | 0.7539 |
|
| 71 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 72 |
|
| 73 |
### Framework versions
|
| 74 |
|
| 75 |
+
- Transformers 5.18.0
|
| 76 |
+
- Pytorch 2.14.1+cu130
|
| 77 |
- Datasets 4.8.5
|
| 78 |
- Tokenizers 0.23.2
|
config.json
CHANGED
|
@@ -34,7 +34,7 @@
|
|
| 34 |
"pad_token_id": 1,
|
| 35 |
"problem_type": "single_label_classification",
|
| 36 |
"tie_word_embeddings": true,
|
| 37 |
-
"transformers_version": "5.
|
| 38 |
"type_vocab_size": 1,
|
| 39 |
"use_cache": true,
|
| 40 |
"vocab_size": 50265
|
|
|
|
| 34 |
"pad_token_id": 1,
|
| 35 |
"problem_type": "single_label_classification",
|
| 36 |
"tie_word_embeddings": true,
|
| 37 |
+
"transformers_version": "5.18.0",
|
| 38 |
"type_vocab_size": 1,
|
| 39 |
"use_cache": true,
|
| 40 |
"vocab_size": 50265
|
emissions.csv
CHANGED
|
@@ -1,2 +1,2 @@
|
|
| 1 |
timestamp,project_name,run_id,experiment_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,water_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,cpu_utilization_percent,gpu_utilization_percent,ram_utilization_percent,ram_used_gb,on_cloud,pue,wue
|
| 2 |
-
2026-
|
|
|
|
| 1 |
timestamp,project_name,run_id,experiment_id,duration,emissions,emissions_rate,cpu_power,gpu_power,ram_power,cpu_energy,gpu_energy,ram_energy,energy_consumed,water_consumed,country_name,country_iso_code,region,cloud_provider,cloud_region,os,python_version,codecarbon_version,cpu_count,cpu_model,gpu_count,gpu_model,longitude,latitude,ram_total_size,tracking_mode,cpu_utilization_percent,gpu_utilization_percent,ram_utilization_percent,ram_used_gb,on_cloud,pue,wue
|
| 2 |
+
2026-10-06T17:03:10,VulnTrain,f97f5062-11d3-48e3-baf8-1137ac1c569f,5b0fa12a-3dd7-45bb-9766-cc326314d9f1,17323.065519523,0.527089473218873,3.042703224927748e-05,70.0002182038376,902.0948631070827,70.0,0.3367934261991617,4.334508493715132,0.33605558670125557,5.007357506615553,0.0,Luxembourg,LUX,luxembourg,,,Linux-6.8.0-146-generic-x86_64-with-glibc2.39,3.12.3,3.3.1,224,Intel(R) Xeon(R) Platinum 8480+,4,4 x NVIDIA L40S,6.1327,49.6098,2015.3354225158691,machine,1.0509953434225843,75.42930733410942,1.0833061699650757,21.249004599823248,N,1.0,0.0
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 498618976
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f41377fab50baa54c54fa7fe391c7485ae1e7c2bdaa2bcbbb942a6e1f58af8f3
|
| 3 |
size 498618976
|