This is a Cross Encoder model finetuned from BAAI/bge-reranker-v2-m3 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: BAAI/bge-reranker-v2-m3
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.

python

from sentence_transformers import CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("pujithapsx/finetuned-bge-reranker-address-25l")
# Get scores for pairs of texts
pairs = [
    ['c/o gupta mg road indore', 'c/o gupta mg road ahmedabad'],
    ['d-101 sector 62 noida', 'd-102 sector 62 noida'],
    ['h.no 45-67 jayanagar bangalore', 'h.no 4567 jayanagar bangalore'],
    ['45 8th main indiranagar bangalore', 'indiranagar 45 8th main bangalore'],
    ['mvp colony visakhapatnam', 'mvp colony hyderabad'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'c/o gupta mg road indore',
    [
        'c/o gupta mg road ahmedabad',
        'd-102 sector 62 noida',
        'h.no 4567 jayanagar bangalore',
        'indiranagar 45 8th main bangalore',
        'mvp colony hyderabad',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Classification

Dataset: entity-matching-eval
Evaluated with CrossEncoderClassificationEvaluator

Metric	Value
accuracy	0.9716
accuracy_threshold	0.0042
f1	0.974
f1_threshold	0.0042
precision	0.9615
recall	0.9868
average_precision	0.984

Training Details

Training Dataset

Unnamed Dataset

Size: 981 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 981 samples:
sentence1 sentence2 label
type string string int
details
min: 8 characters
mean: 27.85 characters
max: 74 characters
min: 8 characters
mean: 27.89 characters
max: 62 characters
0: ~46.48%
1: ~53.52%
Samples:
sentence1 sentence2 label
a-301 royal residency indore a-301 royal res indore 1
kukatpally hyderabad plot 45 plot 45 phase 2 kukatpally hyderabad 1
a-301 royal residency indore a-301 royal res indore 1
Loss: BinaryCrossEntropyLoss with these parameters:
json
1{ 2 "activation_fn": "torch.nn.modules.linear.Identity", 3 "pos_weight": null 4}

Evaluation Dataset

Unnamed Dataset

Size: 141 evaluation samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 141 samples:
sentence1 sentence2 label
type string string int
details
min: 10 characters
mean: 27.66 characters
max: 55 characters
min: 9 characters
mean: 28.26 characters
max: 51 characters
0: ~46.10%
1: ~53.90%
Samples:
sentence1 sentence2 label
c/o gupta mg road indore c/o gupta mg road ahmedabad 0
d-101 sector 62 noida d-102 sector 62 noida 0
h.no 45-67 jayanagar bangalore h.no 4567 jayanagar bangalore 1
Loss: BinaryCrossEntropyLoss with these parameters:
json
1{ 2 "activation_fn": "torch.nn.modules.linear.Identity", 3 "pos_weight": null 4}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_eval_batch_size: 16
learning_rate: 2e-05
weight_decay: 0.01
warmup_steps: 36
remove_unused_columns: False
load_best_model_at_end: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: steps
prediction_loss_only: True
per_device_train_batch_size: 8
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: 2e-05
weight_decay: 0.01
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 3
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_ratio: 0.0
warmup_steps: 36
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: False
label_names: None
load_best_model_at_end: True
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	Validation Loss	entity-matching-eval_average_precision
0.1463	18	0.3776	-	-
0.2927	36	0.1942	0.5548	0.9583
0.4390	54	0.3252	-	-
0.5854	72	0.2161	0.3014	0.9740
0.7317	90	0.4467	-	-
0.8780	108	0.1705	0.1924	0.9899
1.0244	126	0.2846	-	-
1.1707	144	0.143	0.2629	0.9878
1.3171	162	0.1257	-	-
1.4634	180	0.1296	0.3058	0.9818
1.6098	198	0.1998	-	-
1.7561	216	0.0981	0.1660	0.9853
1.9024	234	0.1277	-	-
2.0488	252	0.0216	0.1888	0.9906
2.1951	270	0.1826	-	-
2.3415	288	0.0567	0.2956	0.9594
2.4878	306	0.0929	-	-
2.6341	324	0.0754	0.2090	0.9807
2.7805	342	0.0239	-	-
2.9268	360	0.0268	0.2494	0.9840

Framework Versions

Python: 3.10.12
Sentence Transformers: 5.3.0
Transformers: 4.57.6
PyTorch: 2.10.0+cu128
Accelerate: 1.13.0
Datasets: 4.8.4
Tokenizers: 0.22.2

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",
}

	sentence1	sentence2	label
type	string	string	int
details	min: 8 characters mean: 27.85 characters max: 74 characters	min: 8 characters mean: 27.89 characters max: 62 characters	0: ~46.48% 1: ~53.52%

sentence1	sentence2	label
`a-301 royal residency indore`	`a-301 royal res indore`	`1`
`kukatpally hyderabad plot 45`	`plot 45 phase 2 kukatpally hyderabad`	`1`
`a-301 royal residency indore`	`a-301 royal res indore`	`1`

sentence1	sentence2	label
`c/o gupta mg road indore`	`c/o gupta mg road ahmedabad`	`0`
`d-101 sector 62 noida`	`d-102 sector 62 noida`	`0`
`h.no 45-67 jayanagar bangalore`	`h.no 4567 jayanagar bangalore`	`1`