This is a sentence-transformers model finetuned from BAAI/bge-m3. It maps sentences & paragraphs to a 1024-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: BAAI/bge-m3
Maximum Sequence Length: 512 tokens
Output Dimensionality: 1024 dimensions
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False, 'architecture': 'PeftModelForFeatureExtraction'})
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("DungHugging/sacombank-bge-m3-full")
# Run inference
sentences = [
    'ưu đãi Buffet hải sản giá rẻ tại Saigon Seafood',
    'hồ sơ vay tín chấp được duyệt nhanh qua App trong 1 giờ',
    'Áp dụng tiêu chuẩn CRS (Common Reporting Standard) trong trao đổi thông tin thuế quốc tế.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.4611, 0.4808],
#         [0.4611, 1.0000, 0.6015],
#         [0.4808, 0.6015, 1.0000]])

Evaluation

Metrics

Semantic Similarity

Datasets: spec_sim and final_evaluation
Evaluated with EmbeddingSimilarityEvaluator

Metric	spec_sim	final_evaluation
pearson_cosine	0.5078	0.5078
spearman_cosine	0.5003	0.5003

Binary Classification

Dataset: spec_bin
Evaluated with BinaryClassificationEvaluator

Metric	Value
cosine_accuracy	0.7399
cosine_accuracy_threshold	0.7662
cosine_f1	0.7705
cosine_f1_threshold	0.7137
cosine_precision	0.6714
cosine_recall	0.9038
cosine_ap	0.7812
cosine_mcc	0.452

Training Details

Training Dataset

Unnamed Dataset

Size: 2,665 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: 6 tokens
mean: 14.79 tokens
max: 30 tokens
min: 8 tokens
mean: 18.25 tokens
max: 49 tokens
min: 0.0
mean: 0.5
max: 1.0

Samples:

sentence_0	sentence_1	label
`Lãi suất cơ sở (Base Rate) - dùng làm mốc tham chiếu để tính lãi vay khách hàng.`	`Lãi suất qua đêm (Overnight Rate) - lãi suất vay nóng giữa các ngân hàng trên thị trường liên ngân hàng.`	`0.0`
`Khoản thanh toán lớn vào cuối kỳ hạn vay.`	`Khoản vay có cấu trúc Balloon Payment tại thời điểm đáo hạn.`	`1.0`
`Bảo hiểm trách nhiệm dân sự chủ doanh nghiệp`	`Bảo hiểm tai nạn con người 24/7`	`0.0`

Loss: ContrastiveLoss with these parameters:
json
1{ 2 "distance_metric": "SiameseDistanceMetric.COSINE_DISTANCE", 3 "margin": 0.5, 4 "size_average": true 5}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
per_device_eval_batch_size: 16
num_train_epochs: 10
multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
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: 10
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: round_robin
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss	spec_sim_spearman_cosine	spec_bin_cosine_ap	final_evaluation_spearman_cosine
0.4970	83	-	-0.1802	0.4537	-
0.9940	166	-	-0.0618	0.4928	-
1.0	167	-	-0.0598	0.4935	-
1.4910	249	-	0.0678	0.5431	-
1.9880	332	-	0.1480	0.5821	-
2.0	334	-	0.1500	0.5829	-
2.4850	415	-	0.2448	0.6273	-
2.9820	498	-	0.3185	0.6705	-
2.9940	500	0.0327	-	-	-
3.0	501	-	0.3199	0.6715	-
3.4790	581	-	0.3650	0.7018	-
3.9760	664	-	0.3993	0.7226	-
4.0	668	-	0.3986	0.7222	-
4.4731	747	-	0.4210	0.7354	-
4.9701	830	-	0.4380	0.7469	-
5.0	835	-	0.4375	0.7467	-
5.4671	913	-	0.4514	0.7525	-
5.9641	996	-	0.4606	0.7584	-
5.9880	1000	0.0241	-	-	-
6.0	1002	-	0.4613	0.7591	-
6.4611	1079	-	0.4717	0.7648	-
6.9581	1162	-	0.4791	0.7684	-
7.0	1169	-	0.4799	0.7689	-
7.4551	1245	-	0.4848	0.7712	-
7.9521	1328	-	0.4910	0.7750	-
8.0	1336	-	0.4915	0.7760	-
8.4491	1411	-	0.4955	0.7780	-
8.9461	1494	-	0.4972	0.7796	-
8.9820	1500	0.0217	-	-	-
9.0	1503	-	0.4972	0.7796	-
9.4431	1577	-	0.4995	0.7807	-
9.9401	1660	-	0.5003	0.7812	-
10.0	1670	-	0.5003	0.7812	-
-1	-1	-	-	-	0.5003

Framework Versions

Python: 3.12.12
Sentence Transformers: 5.1.1
Transformers: 4.57.1
PyTorch: 2.8.0+cu126
Accelerate: 1.11.0
Datasets: 4.4.2
Tokenizers: 0.22.1

Citation

BibTeX

Sentence Transformers

bibtex

@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",
}

ContrastiveLoss

bibtex

@inproceedings{hadsell2006dimensionality,
    author={Hadsell, R. and Chopra, S. and LeCun, Y.},
    booktitle={2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06)},
    title={Dimensionality Reduction by Learning an Invariant Mapping},
    year={2006},
    volume={2},
    number={},
    pages={1735-1742},
    doi={10.1109/CVPR.2006.100}
}

	sentence_0	sentence_1	label
type	string	string	float
details	min: 6 tokens mean: 14.79 tokens max: 30 tokens	min: 8 tokens mean: 18.25 tokens max: 49 tokens	min: 0.0 mean: 0.5 max: 1.0