Downloads · 30 days
21
6% of all-time downloads
ubco-mds-2025-capstone-fujitsu-1
ubco-mds-2025-capstone-fujitsu-1/bge-small-en-v1.5-cve-finetuned
bge-small-en-v1.5-cve-finetuned is a sentence similarity model from ubco-mds-2025-capstone-fujitsu-1. Use it when you need a score for how close two texts are. It is set up for sentence-transformers.
This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval.
Downloads · 30 days
21
6% of all-time downloads
All-time downloads
342
Public
Parameters
33.4M
133 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors133 MB · 99%
From the Hugging Face model README
This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for retrieval.
SentenceTransformer(
(0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
(1): Pooling({'embedding_dimension': 384, 'pooling_mode': 'cls', 'include_prompt': True})
(2): Normalize({})
)
First install the Sentence Transformers library:
pip install -U sentence-transformers
Then you can load this model and run inference.
from sentence_transformers import SentenceTransformer
# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
queries = [
'web application behaves unexpectedly when submitting forms',
]
documents = [
'Cross-site scripting (XSS) vulnerability in cgi/client.py in Roundup before 1.4.20 allows remote attackers to inject arbitrary web script or HTML via the @action parameter to support/issue1.',
"Internet Starter, one of SoftCOM iKSORIS system modules, is vulnerable to Reflected XSS (Cross-site Scripting) attacks. An attacker might craft a link containing a malicious script, which then gets directly embedded in references to other resources, what causes the script to run in user's context multiple times.\xa0\nThis vulnerability has been patched in version 79.0",
'XnView Classic for Windows Version 2.40 might allow attackers to cause a denial of service or possibly have unspecified other impact via a crafted .rle file, related to "Data from Faulting Address controls Branch Selection starting at ntdll_77df0000!RtlpCoalesceFreeBlocks+0x00000000000001b6."',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 384] [3, 384]
# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[ 0.4764, 0.2721, -0.1170]])
<!--
### Direct Usage (Transformers)
<details><summary>Click to see the direct usage in Transformers</summary>
</details>
-->
<!--
### Downstream Usage (Sentence Transformers)
You can finetune this model on your own dataset.
<details><summary>Click to expand</summary>
</details>
-->
<!--
### Out-of-Scope Use
*List how the model may foreseeably be misused and address what users ought not to do with the model.*
-->
validation| Metric | Value |
|---|---|
| cosine_accuracy@1 | 0.7662 |
| cosine_accuracy@3 | 0.8624 |
| cosine_accuracy@5 | 0.8904 |
| cosine_accuracy@10 | 0.9229 |
| cosine_precision@1 | 0.7662 |
| cosine_precision@3 | 0.2875 |
| cosine_precision@5 | 0.1781 |
| cosine_precision@10 | 0.0923 |
| cosine_recall@1 | 0.7662 |
| cosine_recall@3 | 0.8624 |
| cosine_recall@5 | 0.8904 |
| cosine_recall@10 | 0.9229 |
| cosine_ndcg@10 | 0.8457 |
| cosine_mrr@10 | 0.8209 |
| cosine_map@100 | 0.8239 |
| sentence_0 | sentence_1 | |
|---|---|---|
| type | string | string |
| modality | text | text |
| details | <ul><li>min: 7 tokens</li><li>mean: 12.11 tokens</li><li>max: 26 tokens</li></ul> | <ul><li>min: 23 tokens</li><li>mean: 97.13 tokens</li><li>max: 512 tokens</li></ul> |
| sentence_0 | sentence_1 |
|---|---|
| <code>Windows Broadcast DVR User Service use after free</code> | <code>Use after free in Windows Broadcast DVR User Service allows an authorized attacker to elevate privileges locally.</code> |
| <code>missing authorization in AnalyticsWP 2.0.0</code> | <code>Missing Authorization vulnerability in Solid Plugins AnalyticsWP allows Accessing Functionality Not Properly Constrained by ACLs.This issue affects AnalyticsWP: from n/a through 2.0.0.</code> |
| <code>authenticated users can access sensitive database information</code> | <code>The Jquery news ticker plugin for WordPress is vulnerable to SQL Injection via the plugin's shortcode in versions up to, and including, 3.0 due to insufficient escaping on the user supplied parameter and lack of sufficient preparation on the existing SQL query. This makes it possible for authenticated attackers with subscriber-level and above permissions to append additional SQL queries into already existing queries that can be used to extract sensitive information from the database.</code> |
{
"scale": 20.0,
"similarity_fct": "cos_sim",
"gather_across_devices": false,
"directions": [
"query_to_doc"
],
"partition_mode": "joint",
"hardness_mode": null,
"hardness_strength": 0.0
}
per_device_train_batch_size: 32num_train_epochs: 1fp16: Trueper_device_eval_batch_size: 32multi_dataset_batch_sampler: round_robinper_device_train_batch_size: 32num_train_epochs: 1max_steps: -1learning_rate: 5e-05lr_scheduler_type: linearlr_scheduler_kwargs: Nonewarmup_steps: 0optim: adamw_torch_fusedoptim_args: Noneweight_decay: 0.0adam_beta1: 0.9adam_beta2: 0.999adam_epsilon: 1e-08optim_target_modules: Nonegradient_accumulation_steps: 1average_tokens_across_devices: Truemax_grad_norm: 1label_smoothing_factor: 0.0bf16: Falsefp16: Truebf16_full_eval: Falsefp16_full_eval: Falsetf32: Nonegradient_checkpointing: Falsegradient_checkpointing_kwargs: Nonetorch_compile: Falsetorch_compile_backend: Nonetorch_compile_mode: Noneuse_liger_kernel: Falseliger_kernel_config: Noneuse_cache: Falseneftune_noise_alpha: Nonetorch_empty_cache_steps: Noneauto_find_batch_size: Falselog_on_each_node: Truelogging_nan_inf_filter: Trueinclude_num_input_tokens_seen: nolog_level: passivelog_level_replica: warningdisable_tqdm: Falseproject: huggingfacetrackio_space_id: Nonetrackio_bucket_id: Nonetrackio_static_space_id: Noneper_device_eval_batch_size: 32prediction_loss_only: Trueeval_on_start: Falseeval_do_concat_batches: Trueeval_use_gather_object: Falseeval_accumulation_steps: Noneinclude_for_metrics: []batch_eval_metrics: Falsesave_only_model: Falsesave_on_each_node: Falseenable_jit_checkpoint: Falsepush_to_hub: Falsehub_private_repo: Nonehub_model_id: Nonehub_strategy: every_savehub_always_push: Falsehub_revision: Noneload_best_model_at_end: Falseignore_data_skip: Falserestore_callback_states_from_checkpoint: Falsefull_determinism: Falseseed: 42data_seed: Noneuse_cpu: Falseaccelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}parallelism_config: Nonedataloader_drop_last: Falsedataloader_num_workers: 0dataloader_pin_memory: Truedataloader_persistent_workers: Falsedataloader_prefetch_factor: Noneremove_unused_columns: Truelabel_names: Nonetrain_sampling_strategy: randomlength_column_name: lengthddp_find_unused_parameters: Noneddp_bucket_cap_mb: Noneddp_broadcast_buffers: Falseddp_static_graph: Noneddp_backend: Noneddp_timeout: 1800fsdp: []fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}deepspeed: Nonedebug: []skip_memory_metrics: Truedo_predict: Falseresume_from_checkpoint: Nonewarmup_ratio: Nonelocal_rank: -1prompts: Nonebatch_sampler: batch_samplermulti_dataset_batch_sampler: round_robinrouter_mapping: {}learning_rate_mapping: {}| Epoch | Step | Training Loss | validation_cosine_ndcg@10 |
|---|---|---|---|
| 0.0186 | 500 | 0.5360 | - |
| 0.0371 | 1000 | 0.2580 | - |
| 0.0557 | 1500 | 0.2020 | - |
| 0.0743 | 2000 | 0.1800 | - |
| 0.0928 | 2500 | 0.1690 | - |
| 0.1114 | 3000 | 0.1648 | - |
| 0.1300 | 3500 | 0.1546 | - |
| 0.1486 | 4000 | 0.1523 | - |
| 0.1671 | 4500 | 0.1444 | - |
| 0.1857 | 5000 | 0.1457 | 0.8294 |
| 0.2043 | 5500 | 0.1441 | - |
| 0.2228 | 6000 | 0.1379 | - |
| 0.2414 | 6500 | 0.1292 | - |
| 0.2600 | 7000 | 0.1260 | - |
| 0.2785 | 7500 | 0.1270 | - |
| 0.2971 | 8000 | 0.1213 | - |
| 0.3157 | 8500 | 0.1234 | - |
| 0.3342 | 9000 | 0.1193 | - |
| 0.3528 | 9500 | 0.1148 | - |
| 0.3714 | 10000 | 0.1213 | 0.8352 |
| 0.3900 | 10500 | 0.1158 | - |
| 0.4085 | 11000 | 0.1183 | - |
| 0.4271 | 11500 | 0.1141 | - |
| 0.4457 | 12000 | 0.1164 | - |
| 0.4642 | 12500 | 0.1047 | - |
| 0.4828 | 13000 | 0.1115 | - |
| 0.5014 | 13500 | 0.1100 | - |
| 0.5199 | 14000 | 0.1084 | - |
| 0.5385 | 14500 | 0.1039 | - |
| 0.5571 | 15000 | 0.1089 | 0.8388 |
| 0.5757 | 15500 | 0.1020 | - |
| 0.5942 | 16000 | 0.1111 | - |
| 0.6128 | 16500 | 0.1124 | - |
| 0.6314 | 17000 | 0.1064 | - |
| 0.6499 | 17500 | 0.1049 | - |
| 0.6685 | 18000 | 0.1032 | - |
| 0.6871 | 18500 | 0.1106 | - |
| 0.7056 | 19000 | 0.1110 | - |
| 0.7242 | 19500 | 0.1023 | - |
| 0.7428 | 20000 | 0.0964 | 0.8434 |
| 0.7613 | 20500 | 0.0957 | - |
| 0.7799 | 21000 | 0.1010 | - |
| 0.7985 | 21500 | 0.0991 | - |
| 0.8171 | 22000 | 0.0983 | - |
| 0.8356 | 22500 | 0.1027 | - |
| 0.8542 | 23000 | 0.0940 | - |
| 0.8728 | 23500 | 0.0967 | - |
| 0.8913 | 24000 | 0.0924 | - |
| 0.9099 | 24500 | 0.1010 | - |
| 0.9285 | 25000 | 0.0961 | 0.8457 |
@inproceedings{reimers-2019-sentence-bert,
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
author = "Reimers, Nils and Gurevych, Iryna",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
month = "11",
year = "2019",
publisher = "Association for Computational Linguistics",
url = "https://arxiv.org/abs/1908.10084",
}
@misc{oord2019representationlearningcontrastivepredictive,
title={Representation Learning with Contrastive Predictive Coding},
author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
year={2019},
eprint={1807.03748},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/1807.03748},
}
<!--
## Glossary
*Clearly define terms in order to be accessible across audiences.*
-->
<!--
## Model Card Authors
*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
-->
<!--
## Model Card Contact
*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
-->