CrossEncoder based on cross-encoder/ms-marco-MiniLM-L12-v2

This is a Cross Encoder model finetuned from cross-encoder/ms-marco-MiniLM-L12-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-L12-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("Davidsamuel101/ft-ms-marco-MiniLM-L12-v2-claims-reranker")
# Get scores for pairs of texts
pairs = [
    ["It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.", 'The report, issued on 18 February 2011, cleared the researchers and "did not find any evidence that NOAA inappropriately manipulated data or failed to adhere to appropriate peer review procedures".'],
    ["It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.", 'Ongoing experiments are conducted by more than 4,000 scientists from many nations.'],
    ["It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.", 'Novell did not seem to proceed to a full court case after losing their case there.'],
    ["It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.", 'In the face of determined opposition from the National Park Service and conservation groups, the dam was never built.'],
    ["It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.", 'At Caltech he developed the first instrument able to measure carbon dioxide in atmospheric samples with consistently reliable accuracy.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    "It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.",
    [
        'The report, issued on 18 February 2011, cleared the researchers and "did not find any evidence that NOAA inappropriately manipulated data or failed to adhere to appropriate peer review procedures".',
        'Ongoing experiments are conducted by more than 4,000 scientists from many nations.',
        'Novell did not seem to proceed to a full court case after losing their case there.',
        'In the face of determined opposition from the National Park Service and conservation groups, the dam was never built.',
        'At Caltech he developed the first instrument able to measure carbon dioxide in atmospheric samples with consistently reliable accuracy.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Training Details

Training Dataset

Unnamed Dataset

Size: 5 training samples
Columns: text1, text2, and label

Approximate statistics based on the first 5 samples:

	text1	text2	label
type	string	string	int
details	min: 186 characters mean: 186.0 characters max: 186 characters	min: 82 characters mean: 122.6 characters max: 197 characters	0: ~80.00% 1: ~20.00%

Samples:

text1	text2	label
`It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.`	`The report, issued on 18 February 2011, cleared the researchers and "did not find any evidence that NOAA inappropriately manipulated data or failed to adhere to appropriate peer review procedures".`	`1`
`It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.`	`Ongoing experiments are conducted by more than 4,000 scientists from many nations.`	`0`
`It found the scientists' rigour and honesty are not in doubt, and their behaviour did not prejudice the IPCC's conclusions, though they did fail to display the proper degree of openness.`	`Novell did not seem to proceed to a full court case after losing their case there.`	`0`

Loss: MultipleNegativesRankingLoss with these parameters:

{
    "scale": 10.0,
    "num_negatives": 4,
    "activation_fn": "torch.nn.modules.activation.Sigmoid"
}

Training Hyperparameters

Non-Default Hyperparameters

eval_strategy: steps
per_device_train_batch_size: 16
learning_rate: 1e-05
num_train_epochs: 10
bf16: True
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: 16
per_device_eval_batch_size: 8
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: 1e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
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
use_ipex: False
bf16: True
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: 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}
tp_size: 0
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}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
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
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: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Framework Versions

Python: 3.13.2
Sentence Transformers: 4.1.0
Transformers: 4.51.3
PyTorch: 2.7.0+cu128
Accelerate: 1.6.0
Datasets: 3.6.0
Tokenizers: 0.21.1

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",
}