End of training

README.md

@@ -9,57 +9,29 @@ metrics:
 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: 0.5002
+- Accuracy: 0.8264
 
 ## Model description
 
-It is a classification model and is aimed to assist in classifying vulnerabilities by severity based on their descriptions.
+More information needed
 
-## How to get started with the model
+## Intended uses & limitations
 
-```python
-from transformers import AutoModelForSequenceClassification, AutoTokenizer
-import torch
+More information needed
 
-labels = ["low", "medium", "high", "critical"]
-
-model_name = "CIRCL/vulnerability-severity-classification-roberta-base"
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForSequenceClassification.from_pretrained(model_name)
-model.eval()
-
-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
 
@@ -74,19 +46,15 @@ The following hyperparameters were used during training:
 - lr_scheduler_type: linear
 - num_epochs: 5
 
-It achieves the following results on the evaluation set:
-- Loss: 0.4987
-- Accuracy: 0.8253
-
 ### Training results
 
 | Training Loss | Epoch | Step   | Validation Loss | Accuracy |
 |:-------------:|:-----:|:------:|:---------------:|:--------:|
-| 0.5635        | 1.0   | 29952  | 0.6258          | 0.7477   |
-| 0.4751        | 2.0   | 59904  | 0.5743          | 0.7755   |
-| 0.4781        | 3.0   | 89856  | 0.5240          | 0.7993   |
-| 0.4643        | 4.0   | 119808 | 0.5056          | 0.8170   |
-| 0.2295        | 5.0   | 149760 | 0.4987          | 0.8253   |
+| 0.6004        | 1.0   | 29952  | 0.6255          | 0.7471   |
+| 0.5196        | 2.0   | 59904  | 0.5728          | 0.7776   |
+| 0.4888        | 3.0   | 89856  | 0.5283          | 0.8019   |
+| 0.3788        | 4.0   | 119808 | 0.5072          | 0.8179   |
+| 0.2821        | 5.0   | 149760 | 0.5002          | 0.8264   |
 
 
 ### Framework versions

emissions.csv

@@ -1,2 +1,2 @@
 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,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,on_cloud,pue
-2025-11-19T14:20:13,codecarbon,81c97158-1708-4a2e-855b-928c690d1130,5b0fa12a-3dd7-45bb-9766-cc326314d9f1,24147.412700983696,0.43390220170688765,1.7968889962658887e-05,42.5,286.31103898999135,94.34468507766725,0.2848920588698454,3.2047893193849575,0.6323957210762677,4.122077099331081,Luxembourg,LUX,luxembourg,,,Linux-6.8.0-71-generic-x86_64-with-glibc2.39,3.12.3,2.8.4,64,AMD EPYC 9124 16-Core Processor,2,2 x NVIDIA L40S,6.1294,49.6113,251.5858268737793,machine,N,1.0
+2025-11-24T18:21:38,codecarbon,a5ed7acf-5313-48b2-824e-d9e8e400e830,5b0fa12a-3dd7-45bb-9766-cc326314d9f1,18042.404894045,0.7567710160236923,4.194402134681441e-05,42.5,360.8287762733843,755.7507977485657,0.2128334216121308,3.1922654485325372,3.7842365660835506,7.189335436228231,Luxembourg,LUX,,,,Linux-6.8.0-88-generic-x86_64-with-glibc2.39,3.12.3,2.8.4,224,Intel(R) Xeon(R) Platinum 8480+,2,2 x NVIDIA H100 NVL,6.1661,49.7498,2015.3354606628418,machine,N,1.0