Upload Cross-Encoder weights (trained on BEDR pairs)

676469e verified 8 days ago

13.4 kB

tags:
  - sentence-transformers
  - cross-encoder
  - reranker
  - generated_from_trainer
  - dataset_size:42906
  - loss:BinaryCrossEntropyLoss
base_model: cross-encoder/ms-marco-MiniLM-L6-v2
pipeline_tag: text-ranking
library_name: sentence-transformers

CrossEncoder based on cross-encoder/ms-marco-MiniLM-L6-v2

This is a Cross Encoder model finetuned from cross-encoder/ms-marco-MiniLM-L6-v2 using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

Model Type: Cross Encoder
Base model: cross-encoder/ms-marco-MiniLM-L6-v2
Maximum Sequence Length: 512 tokens
Number of Output Labels: 1 label

Model Sources

Documentation: Sentence Transformers Documentation
Documentation: Cross Encoder Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Cross Encoders on Hugging Face

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("cross_encoder_model_id")
# Get scores for pairs of texts
pairs = [
    ['TA2541 has used process hollowing to execute CyberGate malware.', 'Woody RAT can create a suspended notepad process and write shellcode to delete a file into the suspended process using `NtWriteVirtualMemory`.'],
    ['Attackers concealed command and control communications by hiding encoded data in the metadata of PNG files.', 'TA551 has hidden encoded data for malware DLLs in a PNG.'],
    ['The cybercriminal created a deceptive Facebook account impersonating a tech support representative to establish initial contact with potential victims.', 'The adversary used klist.exe to view Kerberos ticket expiration times and session keys.'],
    ['APT32 used GetPassword_x64 to harvest credentials.', 'Spyware updated TCC.db to grant full disk access to the malicious application.'],
    ['APT1 uses two utilities, GETMAIL and MAPIGET, to steal email. GETMAIL extracts emails from archived Outlook .pst files.', 'During Frankenstein, the threat actors used MSbuild to execute an actor-created file.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'TA2541 has used process hollowing to execute CyberGate malware.',
    [
        'Woody RAT can create a suspended notepad process and write shellcode to delete a file into the suspended process using `NtWriteVirtualMemory`.',
        'TA551 has hidden encoded data for malware DLLs in a PNG.',
        'The adversary used klist.exe to view Kerberos ticket expiration times and session keys.',
        'Spyware updated TCC.db to grant full disk access to the malicious application.',
        'During Frankenstein, the threat actors used MSbuild to execute an actor-created file.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Training Details

Training Dataset

Unnamed Dataset

Size: 42,906 training samples
Columns: sentence_0, sentence_1, and label

Approximate statistics based on the first 1000 samples:

	sentence_0	sentence_1	label
type	string	string	float
details	min: 32 characters mean: 100.17 characters max: 372 characters	min: 32 characters mean: 100.99 characters max: 403 characters	min: 0.0 mean: 0.48 max: 1.0

Samples:

sentence_0	sentence_1	label
`TA2541 has used process hollowing to execute CyberGate malware.`	`Woody RAT can create a suspended notepad process and write shellcode to delete a file into the suspended process using NtWriteVirtualMemory.`	`1.0`
`Attackers concealed command and control communications by hiding encoded data in the metadata of PNG files.`	`TA551 has hidden encoded data for malware DLLs in a PNG.`	`1.0`
`The cybercriminal created a deceptive Facebook account impersonating a tech support representative to establish initial contact with potential victims.`	`The adversary used klist.exe to view Kerberos ticket expiration times and session keys.`	`0.0`

Loss: BinaryCrossEntropyLoss with these parameters:

{
    "activation_fn": "torch.nn.modules.linear.Identity",
    "pos_weight": null
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 16
per_device_eval_batch_size: 16

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1
num_train_epochs: 3
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
warmup_steps: 0
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
bf16: False
fp16: False
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch_fused
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
project: huggingface
trackio_space_id: trackio
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
hub_revision: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: no
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
liger_kernel_config: None
eval_use_gather_object: False
average_tokens_across_devices: True
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.1864	500	0.6336
0.3729	1000	0.3383
0.5593	1500	0.3197
0.7457	2000	0.3031
0.9321	2500	0.2815
1.1186	3000	0.2555
1.3050	3500	0.2394
1.4914	4000	0.2496
1.6779	4500	0.2312
1.8643	5000	0.2253
2.0507	5500	0.2128
2.2371	6000	0.1872
2.4236	6500	0.2003
2.6100	7000	0.1864
2.7964	7500	0.1992
2.9828	8000	0.1873

Framework Versions

Python: 3.10.19
Sentence Transformers: 5.2.0
Transformers: 4.57.3
PyTorch: 2.9.1
Accelerate: 1.12.0
Datasets: 4.4.2
Tokenizers: 0.22.1

Citation

BibTeX

Sentence Transformers

@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",
}