End of training

README.md

@@ -1,68 +1,50 @@
 ---
 library_name: transformers
-license: cc-by-4.0
+license: mit
 base_model: roberta-base
-metrics:
-- accuracy
 tags:
 - generated_from_trainer
-- text-classification
-- classification
-- nlp
-- vulnerability
+metrics:
+- accuracy
 model-index:
 - name: vulnerability-severity-classification-roberta-base
   results: []
-datasets:
-- CIRCL/vulnerability-scores
 ---
 
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
 
-# VLAI: A RoBERTa-Based Model for Automated Vulnerability Severity Classification
-
-# Severity classification
-
-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).
-
-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)].
-
-**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.
-
-You can read [this page](https://www.vulnerability-lookup.org/user-manual/ai/) for more information.
+# vulnerability-severity-classification-roberta-base
 
+This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.0731
+- Accuracy: 0.8134
+- F1 Macro: 0.7445
+- Low Precision: 0.6507
+- Low Recall: 0.5058
+- Low F1: 0.5692
+- Medium Precision: 0.8479
+- Medium Recall: 0.8626
+- Medium F1: 0.8552
+- High Precision: 0.8069
+- High Recall: 0.8089
+- High F1: 0.8079
+- Critical Precision: 0.7441
+- Critical Recall: 0.7472
+- Critical F1: 0.7457
 
 ## Model description
 
-It is a classification model and is aimed to assist in classifying vulnerabilities by severity based on their descriptions.
-
-## How to get started with the model
-
-```python
-from transformers import AutoModelForSequenceClassification, AutoTokenizer
-import torch
+More information needed
 
-labels = ["low", "medium", "high", "critical"]
+## Intended uses & limitations
 
-model_name = "CIRCL/vulnerability-severity-classification-roberta-base"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSequenceClassification.from_pretrained(model_name)
-model.eval()
+More information needed
 
-test_description = "SAP NetWeaver Visual Composer Metadata Uploader is not protected with a proper authorization, allowing unauthenticated agent to upload potentially malicious executable binaries \
-that could severely harm the host system. This could significantly affect the confidentiality, integrity, and availability of the targeted system."
-inputs = tokenizer(test_description, return_tensors="pt", truncation=True, padding=True)
-
-# Run inference
-with torch.no_grad():
-    outputs = model(**inputs)
-    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
-
-# Print results
-print("Predictions:", predictions)
-predicted_class = torch.argmax(predictions, dim=-1).item()
-print("Predicted severity:", labels[predicted_class])
-```
+## Training and evaluation data
 
+More information needed
 
 ## Training procedure
 
@@ -77,37 +59,20 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: linear
 - num_epochs: 5
 
-It achieves the following results on the evaluation set:
-- Loss: 2.0430
-- Accuracy: 0.8132
-- F1 Macro: 0.7438
-- Low Precision: 0.6379
-- Low Recall: 0.5097
-- Low F1: 0.5666
-- Medium Precision: 0.8494
-- Medium Recall: 0.8632
-- Medium F1: 0.8562
-- High Precision: 0.8038
-- High Recall: 0.8062
-- High F1: 0.8050
-- Critical Precision: 0.7484
-- Critical Recall: 0.7460
-- Critical F1: 0.7472
-
 ### Training results
 
 | 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 |
 |:-------------:|:-----:|:-----:|:---------------:|:--------:|:--------:|:-------------:|:----------:|:------:|:----------------:|:-------------:|:---------:|:--------------:|:-----------:|:-------:|:------------------:|:---------------:|:-----------:|
-| 2.5898        | 1.0   | 15845 | 2.5850          | 0.7385   | 0.6295   | 0.5772        | 0.2855     | 0.3820 | 0.7645           | 0.8514        | 0.8056    | 0.7327         | 0.7009      | 0.7164  | 0.6611             | 0.5731          | 0.6140      |
-| 2.4114        | 2.0   | 31690 | 2.3649          | 0.7600   | 0.6809   | 0.5056        | 0.4649     | 0.4844 | 0.8245           | 0.8086        | 0.8164    | 0.7492         | 0.7480      | 0.7486  | 0.6423             | 0.7097          | 0.6743      |
-| 1.9995        | 3.0   | 47535 | 2.1359          | 0.7862   | 0.7087   | 0.6127        | 0.4467     | 0.5167 | 0.8134           | 0.8654        | 0.8386    | 0.8038         | 0.7379      | 0.7694  | 0.6803             | 0.7424          | 0.7100      |
-| 1.4062        | 4.0   | 63380 | 2.0928          | 0.8063   | 0.7176   | 0.7134        | 0.3661     | 0.4839 | 0.8441           | 0.8591        | 0.8515    | 0.7805         | 0.8172      | 0.7985  | 0.7538             | 0.7201          | 0.7365      |
-| 1.5982        | 5.0   | 79225 | 2.0430          | 0.8132   | 0.7438   | 0.6379        | 0.5097     | 0.5666 | 0.8494           | 0.8632        | 0.8562    | 0.8038         | 0.8062      | 0.8050  | 0.7484             | 0.7460          | 0.7472      |
+| 2.7651        | 1.0   | 15900 | 2.5756          | 0.7357   | 0.6295   | 0.6394        | 0.2587     | 0.3683 | 0.7911           | 0.8066        | 0.7988    | 0.7049         | 0.7369      | 0.7205  | 0.6314             | 0.6294          | 0.6304      |
+| 2.3544        | 2.0   | 31800 | 2.3242          | 0.7640   | 0.6742   | 0.6876        | 0.3302     | 0.4461 | 0.8154           | 0.8253        | 0.8204    | 0.7468         | 0.7512      | 0.7490  | 0.6454             | 0.7212          | 0.6812      |
+| 2.3762        | 3.0   | 47700 | 2.1947          | 0.7844   | 0.7118   | 0.5717        | 0.4992     | 0.5330 | 0.8218           | 0.8440        | 0.8328    | 0.7816         | 0.7704      | 0.7760  | 0.7109             | 0.7001          | 0.7055      |
+| 1.5527        | 4.0   | 63600 | 2.0991          | 0.8034   | 0.7263   | 0.7186        | 0.4157     | 0.5267 | 0.8411           | 0.8543        | 0.8476    | 0.7886         | 0.8028      | 0.7956  | 0.7235             | 0.7472          | 0.7352      |
+| 1.2645        | 5.0   | 79500 | 2.0731          | 0.8134   | 0.7445   | 0.6507        | 0.5058     | 0.5692 | 0.8479           | 0.8626        | 0.8552    | 0.8069         | 0.8089      | 0.8079  | 0.7441             | 0.7472          | 0.7457      |
 
 
 ### Framework versions
 
-- Transformers 5.5.3
+- Transformers 5.5.4
 - Pytorch 2.11.0+cu130
 - Datasets 4.8.4
 - Tokenizers 0.22.2

config.json

@@ -34,7 +34,7 @@
   "pad_token_id": 1,
   "problem_type": "single_label_classification",
   "tie_word_embeddings": true,
-  "transformers_version": "5.5.3",
+  "transformers_version": "5.5.4",
   "type_vocab_size": 1,
   "use_cache": true,
   "vocab_size": 50265

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b605096095a323bc53778e6f78fe1d981595c3bd8d0b622753844e4d80c7358a
+oid sha256:bba16df26f3402e6788784f3f0aef052f24ee0195d123269efceea165997e619
 size 498618952