Sentence Similarity
sentence-transformers
Safetensors
English
modernbert
feature-extraction
cybersecurity
retrieval
mitre-attack
sigma
cve
text-embeddings-inference
Instructions to use alirezaaminzadeh/SecEmbed-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- sentence-transformers
How to use alirezaaminzadeh/SecEmbed-base with sentence-transformers:
from sentence_transformers import SentenceTransformer model = SentenceTransformer("alirezaaminzadeh/SecEmbed-base") sentences = [ "That is a happy person", "That is a happy dog", "That is a very happy person", "Today is a sunny day" ] embeddings = model.encode(sentences) similarities = model.similarity(embeddings, embeddings) print(similarities.shape) # [4, 4] - Notebooks
- Google Colab
- Kaggle
Add evaluation metrics
Browse files- eval_metrics.json +68 -0
eval_metrics.json
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"baseline": {
|
| 3 |
+
"attack_retrieval": {
|
| 4 |
+
"recall@5": 0.05,
|
| 5 |
+
"recall@10": 0.1,
|
| 6 |
+
"n": 40
|
| 7 |
+
},
|
| 8 |
+
"cve_similarity": {
|
| 9 |
+
"recall@5": 0.325,
|
| 10 |
+
"recall@10": 0.4,
|
| 11 |
+
"n": 40
|
| 12 |
+
},
|
| 13 |
+
"cwe_retrieval": {
|
| 14 |
+
"recall@5": 0.4,
|
| 15 |
+
"recall@10": 0.475,
|
| 16 |
+
"n": 40
|
| 17 |
+
},
|
| 18 |
+
"sigma_retrieval": {
|
| 19 |
+
"recall@5": 0.0,
|
| 20 |
+
"recall@10": 0.0,
|
| 21 |
+
"n": 40
|
| 22 |
+
},
|
| 23 |
+
"soc_playbook": {
|
| 24 |
+
"recall@5": 0.05,
|
| 25 |
+
"recall@10": 0.125,
|
| 26 |
+
"n": 40
|
| 27 |
+
},
|
| 28 |
+
"threat_report_retrieval": {
|
| 29 |
+
"recall@5": 0.025,
|
| 30 |
+
"recall@10": 0.125,
|
| 31 |
+
"n": 40
|
| 32 |
+
},
|
| 33 |
+
"macro_recall@5": 0.1417
|
| 34 |
+
},
|
| 35 |
+
"final": {
|
| 36 |
+
"attack_retrieval": {
|
| 37 |
+
"recall@5": 1.0,
|
| 38 |
+
"recall@10": 1.0,
|
| 39 |
+
"n": 80
|
| 40 |
+
},
|
| 41 |
+
"cve_similarity": {
|
| 42 |
+
"recall@5": 0.9625,
|
| 43 |
+
"recall@10": 0.9625,
|
| 44 |
+
"n": 80
|
| 45 |
+
},
|
| 46 |
+
"cwe_retrieval": {
|
| 47 |
+
"recall@5": 1.0,
|
| 48 |
+
"recall@10": 1.0,
|
| 49 |
+
"n": 80
|
| 50 |
+
},
|
| 51 |
+
"sigma_retrieval": {
|
| 52 |
+
"recall@5": 0.8875,
|
| 53 |
+
"recall@10": 0.9,
|
| 54 |
+
"n": 80
|
| 55 |
+
},
|
| 56 |
+
"soc_playbook": {
|
| 57 |
+
"recall@5": 1.0,
|
| 58 |
+
"recall@10": 1.0,
|
| 59 |
+
"n": 80
|
| 60 |
+
},
|
| 61 |
+
"threat_report_retrieval": {
|
| 62 |
+
"recall@5": 0.975,
|
| 63 |
+
"recall@10": 0.975,
|
| 64 |
+
"n": 80
|
| 65 |
+
},
|
| 66 |
+
"macro_recall@5": 0.9708
|
| 67 |
+
}
|
| 68 |
+
}
|