xlm-roberta-ua-distilled

SentenceTransformer based on FacebookAI/xlm-roberta-base

This is a sentence-transformers model finetuned from FacebookAI/xlm-roberta-base. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

👉 Check out the model on GitHub.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: FacebookAI/xlm-roberta-base <!-- at revision e73636d4f797dec63c3081bb6ed5c7b0bb3f2089 -->
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset: parallel-sentences-talks, parallel-sentences-wikimatrix, parallel-sentences-tatoeba
• Language: Ukrainian, English
• License: MIT

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}) with Transformer model: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

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 SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("panalexeu/xlm-roberta-ua-distilled")
# Run inference
sentences = [
    "You'd better consult the doctor.",
    'Краще проконсультуйся у лікаря.',
    'Їх позначають як Aufklärungsfahrzeug 93 та Aufklärungsfahrzeug 97 відповідно.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

<!-- ### 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.* -->

Evaluation

Metrics

Knowledge Distillation

• Dataset: mse-en-ua
• Evaluated with <code>MSEEvaluator</code>

Semantic Similarity

• Datasets: sts17-en-en, sts17-en-ua and sts17-ua-ua
• Evaluated with <code>EmbeddingSimilarityEvaluator</code>

<!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

• Dataset: parallel-sentences-talks, parallel-sentences-wikimatrix, parallel-sentences-tatoeba
• Size: 523,982 training samples
• Columns: <code>english</code>, <code>non_english</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	<ul><li>min: 5 tokens</li><li>mean: 21.11 tokens</li><li>max: 254 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 23.15 tokens</li><li>max: 293 tokens</li></ul>	<ul><li>size: 768 elements</li></ul>

• Samples:

  english	non_english	label
  <code>Her real name is Lydia (リディア, Ridia), but she was mistaken for a boy and called Ricard.</code>	<code>Справжнє ім'я — Лідія, але її помилково сприйняли за хлопчика і назвали Рікард.</code>	<code>[0.15217968821525574, -0.17830222845077515, -0.12677159905433655, 0.22082313895225525, 0.40085524320602417, ...]</code>
  <code>(Applause) So he didn't just learn water.</code>	<code>(Аплодисменти) Він не тільки вивчив слово "вода".</code>	<code>[-0.1058148592710495, -0.08846072107553482, -0.2684604823589325, -0.105219267308712, 0.3050258755683899, ...]</code>
  <code>It is tightly integrated with SAM, the Storage and Archive Manager, and hence is often referred to as SAM-QFS.</code>	<code>Вона тісно інтегрована з SAM (Storage and Archive Manager), тому часто називається SAM-QFS.</code>	<code>[0.03270340710878372, -0.45798248052597046, -0.20090211927890778, 0.006579531356692314, -0.03178019821643829, ...]</code>

• Loss: <code>MSELoss</code>

Evaluation Dataset

• Dataset: parallel-sentences-talks, parallel-sentences-wikimatrix, parallel-sentences-tatoeba
• Size: 3,838 evaluation samples
• Columns: <code>english</code>, <code>non_english</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	<ul><li>min: 5 tokens</li><li>mean: 15.64 tokens</li><li>max: 143 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 16.98 tokens</li><li>max: 148 tokens</li></ul>	<ul><li>size: 768 elements</li></ul>

• Samples:

  english	non_english	label
  <code>I have lost my wallet.</code>	<code>Я загубив гаманець.</code>	<code>[-0.11186987161636353, -0.03419225662946701, -0.31304317712783813, 0.0838347002863884, 0.108644500374794, ...]</code>
  <code>It's a pharmaceutical product.</code>	<code>Це фармацевтичний продукт.</code>	<code>[0.04133488982915878, -0.4182000756263733, -0.30786487460136414, -0.09351564198732376, -0.023946482688188553, ...]</code>
  <code>We've all heard of the Casual Friday thing.</code>	<code>Всі ми чули про «джинсову п’ятницю» (вільна форма одягу).</code>	<code>[-0.10697802156209946, 0.21002227067947388, -0.2513434886932373, -0.3718843460083008, 0.06871984899044037, ...]</code>

• Loss: <code>MSELoss</code>

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• gradient_accumulation_steps: 3
• num_train_epochs: 4
• warmup_ratio: 0.1

All Hyperparameters

<details><summary>Click to expand</summary>

• 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: 3
• 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: 1.0
• num_train_epochs: 4
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• 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: 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}
• 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

</details>

Training Logs

Framework Versions

• Python: 3.11.11
• Sentence Transformers: 3.4.1
• Transformers: 4.51.1
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.5.0
• Tokenizers: 0.21.0

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",
}

MSELoss

@inproceedings{reimers-2020-multilingual-sentence-bert,
    title = "Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2020",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/2004.09813",
}

<!-- ## 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.* -->