README.md

metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:28783032
  - loss:MSELoss
base_model: BAAI/bge-small-en-v1.5
widget:
  - source_sentence: It is 98 percent employee-owned.
    sentences:
      - >-
        Nos encontramos en una situación histórica en la que estamos negociando
        con 12 Estados miembros o vamos a iniciar negociaciones y ante situación
        en la que el crecimiento conjunto de Europa se puede hacer realidad en
        un tiempo previsible.
      - >-
        OBESSU es formalmente reconocido por el Consejo de Europa y la Unión
        europea y es un regular interlocutor de la Comisión europea, Eurocámara,
        Consejo de Europa y UNESCO.
      - Son el 98% de las empleadas domésticas.
  - source_sentence: >-
      There is the continuing situation with regard to Article 235, as the
      honourable Member has mentioned, and I have to say that Article 235 is the
      only possible legal base for the actions that are carried out under the
      current Helios programme.
    sentences:
      - >-
        Una investigación inicial en el museo mostró "fallas flagrantes" en la
        seguridad, donde funcionaban apropiadamente sólo siete de las 43 cámaras
        instaladas, informaron medios estatales.
      - >-
        La votación de mañana también es importante para nosotros porque hemos
        presentado una serie de enmiendas que hacen hincapié en la preferencia
        comunitaria.
      - >-
        Continúa la situación en lo que respecta al artículo 235, como ha
        mencionado su Señoría, y debo decir que dicho artículo 235 es el único
        fundamento jurídico posible para emprender las acciones que se llevan a
        cabo bajo el actual programa Helios.
  - source_sentence: People Finding Gold And Coal At Mining Work
    sentences:
      - Se emocionaron al recibir a tantos visitantes de diferentes países.
      - encontrar oro y carbón en el trabajo minero personas
      - '- ¿Está el fax preparado?'
  - source_sentence: Come on, darling.
    sentences:
      - Vamos, cariño.
      - >-
        – Señor Presidente, por lo que se refiere a la enmienda 15, desearía que
        se considerase como un añadido.
      - Digger más tarde apareció en Superboy con otro escuadrón.
  - source_sentence: Why are you so afraid?
    sentences:
      - >-
        Segundos después, un asaltante enmascarado vuela la puerta de la
        habitación con explosivos, mata sus asesores y luego procede a
        secuestrar a Ashton.
      - '- ¿Roma desde mi teléfono?'
      - ¿Por qué tienes tanto miedo?
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - negative_mse
  - src2trg_accuracy
  - trg2src_accuracy
  - mean_accuracy
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: SentenceTransformer based on BAAI/bge-small-en-v1.5
    results:
      - task:
          type: knowledge-distillation
          name: Knowledge Distillation
        dataset:
          name: mse evaluation en es
          type: mse_evaluation-en-es
        metrics:
          - type: negative_mse
            value: -0.046344783157110214
            name: Negative Mse
      - task:
          type: translation
          name: Translation
        dataset:
          name: translation evaluation en es
          type: translation_evaluation-en-es
        metrics:
          - type: src2trg_accuracy
            value: 0.9417
            name: Src2Trg Accuracy
          - type: trg2src_accuracy
            value: 0.9294
            name: Trg2Src Accuracy
          - type: mean_accuracy
            value: 0.93555
            name: Mean Accuracy
      - task:
          type: knowledge-distillation
          name: Knowledge Distillation
        dataset:
          name: mse evaluation en pt
          type: mse_evaluation-en-pt
        metrics:
          - type: negative_mse
            value: -0.04470514878630638
            name: Negative Mse
      - task:
          type: translation
          name: Translation
        dataset:
          name: translation evaluation en pt
          type: translation_evaluation-en-pt
        metrics:
          - type: src2trg_accuracy
            value: 0.9513
            name: Src2Trg Accuracy
          - type: trg2src_accuracy
            value: 0.9384
            name: Trg2Src Accuracy
          - type: mean_accuracy
            value: 0.94485
            name: Mean Accuracy
      - task:
          type: knowledge-distillation
          name: Knowledge Distillation
        dataset:
          name: mse evaluation en pt br
          type: mse_evaluation-en-pt-br
        metrics:
          - type: negative_mse
            value: -0.039737459272146225
            name: Negative Mse
      - task:
          type: translation
          name: Translation
        dataset:
          name: translation evaluation en pt br
          type: translation_evaluation-en-pt-br
        metrics:
          - type: src2trg_accuracy
            value: 0.9789
            name: Src2Trg Accuracy
          - type: trg2src_accuracy
            value: 0.9728
            name: Trg2Src Accuracy
          - type: mean_accuracy
            value: 0.97585
            name: Mean Accuracy
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: sts17 es en
          type: sts17-es-en
        metrics:
          - type: pearson_cosine
            value: 0.8102009718548969
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.817051723930442
            name: Spearman Cosine

SentenceTransformer based on BAAI/bge-small-en-v1.5

This is a sentence-transformers model finetuned from BAAI/bge-small-en-v1.5. It maps sentences & paragraphs to a 384-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-small-en-v1.5
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 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': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, '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.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
sentences = [
    'Why are you so afraid?',
    '¿Por qué tienes tanto miedo?',
    'Segundos después, un asaltante enmascarado vuela la puerta de la habitación con explosivos, mata sus asesores y luego procede a secuestrar a Ashton.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

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

Evaluation

Metrics

Knowledge Distillation

• Datasets: mse_evaluation-en-es, mse_evaluation-en-pt and mse_evaluation-en-pt-br
• Evaluated with MSEEvaluator

Metric	mse_evaluation-en-es	mse_evaluation-en-pt	mse_evaluation-en-pt-br
negative_mse	-0.0463	-0.0447	-0.0397

Translation

• Datasets: translation_evaluation-en-es, translation_evaluation-en-pt and translation_evaluation-en-pt-br
• Evaluated with TranslationEvaluator

Metric	translation_evaluation-en-es	translation_evaluation-en-pt	translation_evaluation-en-pt-br
src2trg_accuracy	0.9417	0.9513	0.9789
trg2src_accuracy	0.9294	0.9384	0.9728
mean_accuracy	0.9355	0.9448	0.9758

Semantic Similarity

• Dataset: sts17-es-en
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.8102
spearman_cosine	0.8171

Training Details

Training Dataset

Unnamed Dataset

• Size: 28,783,032 training samples
• Columns: english, non_english, and label
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	min: 4 tokens mean: 26.03 tokens max: 131 tokens	min: 4 tokens mean: 41.0 tokens max: 219 tokens	size: 384 elements

• Samples:

  english	non_english	label
  Unlike last year, the eye of the storm this year is Sindh where flood waters have razed in numerous villages and displaced millions.	A diferencia del año pasado, el ojo de la tormenta este año está en Sindh , donde las aguas de las inundaciones han arrasado numerosas aldeas y desplazado a millones.	[-0.007684561889618635, 0.008107933215796947, -0.011652171611785889, 0.025253329426050186, 0.03101799078285694, ...]
  Spotted taxis to Jeradda.	Vi taxis a Jeradda.	[0.02564830146729946, -0.0010036780731752515, 0.04843117669224739, -0.002532949671149254, 0.01702607050538063, ...]
  The page has also inspired several other copycat pages on Facebook with similar titles	Asimismo, la página ha inspirado otras páginas imitadoras en Facebook con títulos similares.	[-0.04195510596036911, -0.026423752307891846, -0.08573361486196518, 0.0040077404119074345, -0.02051585540175438, ...]

• Loss: MSELoss

Evaluation Dataset

Unnamed Dataset

• Size: 30,000 evaluation samples
• Columns: english, non_english, and label
• Approximate statistics based on the first 1000 samples:

  	english	non_english	label
  type	string	string	list
  details	min: 3 tokens mean: 22.91 tokens max: 104 tokens	min: 3 tokens mean: 36.22 tokens max: 203 tokens	size: 384 elements

• Samples:

  english	non_english	label
  It goes with your hair.	Va bien con su pelo.	[0.013736451044678688, -0.014708670787513256, 0.020351335406303406, -0.02381581813097, -0.013853796757757664, ...]
  We know that criminality is not limited to our continent, but consists of worldwide networks.	Sabemos que la criminalidad no está limitada a nuestro continente, sino que consta de redes mundiales.	[0.03908732160925865, 0.0009072094107978046, 0.0007832577684894204, -0.044317133724689484, 0.09392903745174408, ...]
  Mm-hm. All except the biggest one.	A todos, excepto al más grande.	[-0.024451246485114098, -0.03345730900764465, -0.04918806627392769, -0.05197983607649803, -0.03147919476032257, ...]

• Loss: MSELoss

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• gradient_accumulation_steps: 2
• learning_rate: 0.0003
• num_train_epochs: 8
• warmup_ratio: 0.15
• bf16: True
• dataloader_num_workers: 8

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 512
• per_device_eval_batch_size: 512
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 2
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 0.0003
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 8
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.15
• 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: 8
• 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}
• 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
• dispatch_batches: None
• split_batches: 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

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	mse_evaluation-en-es_negative_mse	translation_evaluation-en-es_mean_accuracy	mse_evaluation-en-pt_negative_mse	translation_evaluation-en-pt_mean_accuracy	mse_evaluation-en-pt-br_negative_mse	translation_evaluation-en-pt-br_mean_accuracy	sts17-es-en_spearman_cosine
0.0178	500	0.0009	-	-	-	-	-	-	-	-
0.0356	1000	0.0008	-	-	-	-	-	-	-	-
0.0534	1500	0.0007	-	-	-	-	-	-	-	-
0.0712	2000	0.0007	0.0005	-0.10122959	0.6481	-0.10340506	0.6205	-0.10725543	0.5905	0.3595
0.0889	2500	0.0006	-	-	-	-	-	-	-	-
0.1067	3000	0.0006	-	-	-	-	-	-	-	-
0.1245	3500	0.0006	-	-	-	-	-	-	-	-
0.1423	4000	0.0006	0.0004	-0.08365556	0.7992	-0.08426243	0.8051	-0.08419664	0.8281	0.4937
0.1601	4500	0.0005	-	-	-	-	-	-	-	-
0.1779	5000	0.0005	-	-	-	-	-	-	-	-
0.1957	5500	0.0005	-	-	-	-	-	-	-	-
0.2135	6000	0.0005	0.0004	-0.07376316	0.8593	-0.073814325	0.8695	-0.07152441	0.9081	0.6214
0.2312	6500	0.0005	-	-	-	-	-	-	-	-
0.2490	7000	0.0005	-	-	-	-	-	-	-	-
0.2668	7500	0.0005	-	-	-	-	-	-	-	-
0.2846	8000	0.0005	0.0004	-0.06761765	0.8821	-0.067249514	0.8938	-0.06428356	0.9331	0.6909
0.3024	8500	0.0005	-	-	-	-	-	-	-	-
0.3202	9000	0.0005	-	-	-	-	-	-	-	-
0.3380	9500	0.0004	-	-	-	-	-	-	-	-
0.3558	10000	0.0004	0.0003	-0.06365564	0.8960	-0.06300899	0.9072	-0.059927516	0.9428	0.7220
0.3736	10500	0.0004	-	-	-	-	-	-	-	-
0.3913	11000	0.0004	-	-	-	-	-	-	-	-
0.4091	11500	0.0004	-	-	-	-	-	-	-	-
0.4269	12000	0.0004	0.0003	-0.060988236	0.905	-0.060110345	0.9161	-0.0567778	0.9517	0.7385
0.4447	12500	0.0004	-	-	-	-	-	-	-	-
0.4625	13000	0.0004	-	-	-	-	-	-	-	-
0.4803	13500	0.0004	-	-	-	-	-	-	-	-
0.4981	14000	0.0004	0.0003	-0.05906638	0.9109	-0.0581316	0.9222	-0.05460985	0.9569	0.7502
0.5159	14500	0.0004	-	-	-	-	-	-	-	-
0.5336	15000	0.0004	-	-	-	-	-	-	-	-
0.5514	15500	0.0004	-	-	-	-	-	-	-	-
0.5692	16000	0.0004	0.0003	-0.057560425	0.9152	-0.05654549	0.9244	-0.052867033	0.9596	0.7595
0.5870	16500	0.0004	-	-	-	-	-	-	-	-
0.6048	17000	0.0004	-	-	-	-	-	-	-	-
0.6226	17500	0.0004	-	-	-	-	-	-	-	-
0.6404	18000	0.0004	0.0003	-0.05640703	0.9160	-0.055421203	0.9271	-0.051537633	0.9621	0.7660
0.6582	18500	0.0004	-	-	-	-	-	-	-	-
0.6760	19000	0.0004	-	-	-	-	-	-	-	-
0.6937	19500	0.0004	-	-	-	-	-	-	-	-
0.7115	20000	0.0004	0.0003	-0.055573538	0.9194	-0.054422	0.9296	-0.050369795	0.9634	0.7793
0.7293	20500	0.0004	-	-	-	-	-	-	-	-
0.7471	21000	0.0004	-	-	-	-	-	-	-	-
0.7649	21500	0.0004	-	-	-	-	-	-	-	-
0.7827	22000	0.0004	0.0003	-0.054716494	0.9215	-0.05353505	0.9318	-0.049468324	0.9652	0.7807
0.8005	22500	0.0004	-	-	-	-	-	-	-	-
0.8183	23000	0.0004	-	-	-	-	-	-	-	-
0.8360	23500	0.0004	-	-	-	-	-	-	-	-
0.8538	24000	0.0004	0.0003	-0.054093935	0.9230	-0.052881736	0.9325	-0.04880123	0.9662	0.7913
0.8716	24500	0.0004	-	-	-	-	-	-	-	-
0.8894	25000	0.0004	-	-	-	-	-	-	-	-
0.9072	25500	0.0004	-	-	-	-	-	-	-	-
0.9250	26000	0.0004	0.0003	-0.05346901	0.9239	-0.0522654	0.9354	-0.048117638	0.9675	0.7909
0.9428	26500	0.0004	-	-	-	-	-	-	-	-
0.9606	27000	0.0004	-	-	-	-	-	-	-	-
0.9784	27500	0.0004	-	-	-	-	-	-	-	-
0.9961	28000	0.0004	0.0003	-0.05307414	0.9252	-0.051890973	0.9358	-0.047557026	0.968	0.7980
1.0139	28500	0.0004	-	-	-	-	-	-	-	-
1.0317	29000	0.0004	-	-	-	-	-	-	-	-
1.0495	29500	0.0004	-	-	-	-	-	-	-	-
1.0673	30000	0.0004	0.0003	-0.052725878	0.9255	-0.051456288	0.9363	-0.04720709	0.9678	0.7961
1.0851	30500	0.0004	-	-	-	-	-	-	-	-
1.1029	31000	0.0004	-	-	-	-	-	-	-	-
1.1206	31500	0.0004	-	-	-	-	-	-	-	-
1.1384	32000	0.0004	0.0003	-0.05232365	0.9254	-0.051122643	0.9365	-0.046895616	0.9697	0.7980
1.1562	32500	0.0004	-	-	-	-	-	-	-	-
1.1740	33000	0.0004	-	-	-	-	-	-	-	-
1.1918	33500	0.0004	-	-	-	-	-	-	-	-
1.2096	34000	0.0004	0.0003	-0.052056298	0.9265	-0.050777122	0.9382	-0.04632004	0.9692	0.8033
1.2274	34500	0.0004	-	-	-	-	-	-	-	-
1.2452	35000	0.0004	-	-	-	-	-	-	-	-
1.2629	35500	0.0004	-	-	-	-	-	-	-	-
1.2807	36000	0.0004	0.0003	-0.051708605	0.9264	-0.050396774	0.9375	-0.04602845	0.9700	0.8045
1.2985	36500	0.0004	-	-	-	-	-	-	-	-
1.3163	37000	0.0004	-	-	-	-	-	-	-	-
1.3341	37500	0.0004	-	-	-	-	-	-	-	-
1.3519	38000	0.0003	0.0003	-0.05145791	0.9271	-0.050056294	0.9383	-0.045601957	0.9699	0.8009
1.3697	38500	0.0003	-	-	-	-	-	-	-	-
1.3875	39000	0.0003	-	-	-	-	-	-	-	-
1.4053	39500	0.0003	-	-	-	-	-	-	-	-
1.4230	40000	0.0003	0.0003	-0.05112224	0.9286	-0.049697492	0.9385	-0.04513698	0.9706	0.8026
1.4408	40500	0.0003	-	-	-	-	-	-	-	-
1.4586	41000	0.0003	-	-	-	-	-	-	-	-
1.4764	41500	0.0003	-	-	-	-	-	-	-	-
1.4942	42000	0.0003	0.0003	-0.05098933	0.9277	-0.04952781	0.9397	-0.045018516	0.9706	0.8064
1.5120	42500	0.0003	-	-	-	-	-	-	-	-
1.5298	43000	0.0003	-	-	-	-	-	-	-	-
1.5476	43500	0.0003	-	-	-	-	-	-	-	-
1.5653	44000	0.0003	0.0003	-0.05063988	0.9300	-0.04921307	0.9405	-0.044630885	0.971	0.8053
1.5831	44500	0.0003	-	-	-	-	-	-	-	-
1.6009	45000	0.0003	-	-	-	-	-	-	-	-
1.6187	45500	0.0003	-	-	-	-	-	-	-	-
1.6365	46000	0.0003	0.0003	-0.050518364	0.9292	-0.04895482	0.9400	-0.044507742	0.9717	0.8025
1.6543	46500	0.0003	-	-	-	-	-	-	-	-
1.6721	47000	0.0003	-	-	-	-	-	-	-	-
1.6899	47500	0.0003	-	-	-	-	-	-	-	-
1.7077	48000	0.0003	0.0003	-0.05039251	0.9295	-0.048876137	0.9402	-0.044356253	0.9717	0.8000
1.7254	48500	0.0003	-	-	-	-	-	-	-	-
1.7432	49000	0.0003	-	-	-	-	-	-	-	-
1.7610	49500	0.0003	-	-	-	-	-	-	-	-
1.7788	50000	0.0003	0.0003	-0.050093956	0.9299	-0.04855838	0.9405	-0.044041798	0.9715	0.8029
1.7966	50500	0.0003	-	-	-	-	-	-	-	-
1.8144	51000	0.0003	-	-	-	-	-	-	-	-
1.8322	51500	0.0003	-	-	-	-	-	-	-	-
1.8500	52000	0.0003	0.0003	-0.04991082	0.9303	-0.04846586	0.9408	-0.043873988	0.972	0.8081
1.8677	52500	0.0003	-	-	-	-	-	-	-	-
1.8855	53000	0.0003	-	-	-	-	-	-	-	-
1.9033	53500	0.0003	-	-	-	-	-	-	-	-
1.9211	54000	0.0003	0.0003	-0.049736623	0.9308	-0.048238307	0.9412	-0.04365188	0.9719	0.8024
1.9389	54500	0.0003	-	-	-	-	-	-	-	-
1.9567	55000	0.0003	-	-	-	-	-	-	-	-
1.9745	55500	0.0003	-	-	-	-	-	-	-	-
1.9923	56000	0.0003	0.0003	-0.049630426	0.9304	-0.048146106	0.9409	-0.043522894	0.9724	0.8083
2.0100	56500	0.0003	-	-	-	-	-	-	-	-
2.0278	57000	0.0003	-	-	-	-	-	-	-	-
2.0456	57500	0.0003	-	-	-	-	-	-	-	-
2.0634	58000	0.0003	0.0003	-0.04953543	0.9315	-0.04801518	0.9408	-0.04336739	0.9729	0.8088
2.0812	58500	0.0003	-	-	-	-	-	-	-	-
2.0990	59000	0.0003	-	-	-	-	-	-	-	-
2.1168	59500	0.0003	-	-	-	-	-	-	-	-
2.1346	60000	0.0003	0.0003	-0.049395695	0.9298	-0.047923535	0.9415	-0.04315244	0.9718	0.8021
2.1523	60500	0.0003	-	-	-	-	-	-	-	-
2.1701	61000	0.0003	-	-	-	-	-	-	-	-
2.1879	61500	0.0003	-	-	-	-	-	-	-	-
2.2057	62000	0.0003	0.0003	-0.04926271	0.9308	-0.04775018	0.9415	-0.043019187	0.9728	0.8044
2.2235	62500	0.0003	-	-	-	-	-	-	-	-
2.2413	63000	0.0003	-	-	-	-	-	-	-	-
2.2591	63500	0.0003	-	-	-	-	-	-	-	-
2.2769	64000	0.0003	0.0003	-0.049204785	0.9310	-0.047680687	0.9414	-0.04296754	0.9726	0.8056
2.2946	64500	0.0003	-	-	-	-	-	-	-	-
2.3124	65000	0.0003	-	-	-	-	-	-	-	-
2.3302	65500	0.0003	-	-	-	-	-	-	-	-
2.3480	66000	0.0003	0.0003	-0.049087144	0.9310	-0.04754995	0.9413	-0.042845335	0.9732	0.8008
2.3658	66500	0.0003	-	-	-	-	-	-	-	-
2.3836	67000	0.0003	-	-	-	-	-	-	-	-
2.4014	67500	0.0003	-	-	-	-	-	-	-	-
2.4192	68000	0.0003	0.0003	-0.049001597	0.9308	-0.04741441	0.9419	-0.042673677	0.9733	0.8080
2.4369	68500	0.0003	-	-	-	-	-	-	-	-
2.4547	69000	0.0003	-	-	-	-	-	-	-	-
2.4725	69500	0.0003	-	-	-	-	-	-	-	-
2.4903	70000	0.0003	0.0003	-0.048882827	0.9321	-0.047341634	0.9422	-0.042661835	0.9734	0.8087
2.5081	70500	0.0003	-	-	-	-	-	-	-	-
2.5259	71000	0.0003	-	-	-	-	-	-	-	-
2.5437	71500	0.0003	-	-	-	-	-	-	-	-
2.5615	72000	0.0003	0.0003	-0.048752587	0.9314	-0.04728313	0.9415	-0.042584546	0.9734	0.8069
2.5793	72500	0.0003	-	-	-	-	-	-	-	-
2.5970	73000	0.0003	-	-	-	-	-	-	-	-
2.6148	73500	0.0003	-	-	-	-	-	-	-	-
2.6326	74000	0.0003	0.0003	-0.04871502	0.9319	-0.047124594	0.9424	-0.04228281	0.9733	0.8082
2.6504	74500	0.0003	-	-	-	-	-	-	-	-
2.6682	75000	0.0003	-	-	-	-	-	-	-	-
2.6860	75500	0.0003	-	-	-	-	-	-	-	-
2.7038	76000	0.0003	0.0003	-0.04866153	0.9320	-0.047041744	0.9432	-0.04232903	0.9736	0.8078
2.7216	76500	0.0003	-	-	-	-	-	-	-	-
2.7393	77000	0.0003	-	-	-	-	-	-	-	-
2.7571	77500	0.0003	-	-	-	-	-	-	-	-
2.7749	78000	0.0003	0.0003	-0.04851333	0.9323	-0.046925455	0.9424	-0.04223377	0.9735	0.8134
2.7927	78500	0.0003	-	-	-	-	-	-	-	-
2.8105	79000	0.0003	-	-	-	-	-	-	-	-
2.8283	79500	0.0003	-	-	-	-	-	-	-	-
2.8461	80000	0.0003	0.0003	-0.048524972	0.9329	-0.04693048	0.9425	-0.042133827	0.9725	0.8085
2.8639	80500	0.0003	-	-	-	-	-	-	-	-
2.8817	81000	0.0003	-	-	-	-	-	-	-	-
2.8994	81500	0.0003	-	-	-	-	-	-	-	-
2.9172	82000	0.0003	0.0003	-0.048406757	0.9336	-0.04686156	0.9433	-0.042108946	0.9736	0.8098
2.9350	82500	0.0003	-	-	-	-	-	-	-	-
2.9528	83000	0.0003	-	-	-	-	-	-	-	-
2.9706	83500	0.0003	-	-	-	-	-	-	-	-
2.9884	84000	0.0003	0.0003	-0.048344385	0.9317	-0.046766046	0.9432	-0.042005215	0.9736	0.8109
3.0062	84500	0.0003	-	-	-	-	-	-	-	-
3.0239	85000	0.0003	-	-	-	-	-	-	-	-
3.0417	85500	0.0003	-	-	-	-	-	-	-	-
3.0595	86000	0.0003	0.0003	-0.048280284	0.9314	-0.04669383	0.9431	-0.041867185	0.9738	0.8065
3.0773	86500	0.0003	-	-	-	-	-	-	-	-
3.0951	87000	0.0003	-	-	-	-	-	-	-	-
3.1129	87500	0.0003	-	-	-	-	-	-	-	-
3.1307	88000	0.0003	0.0003	-0.04822296	0.9325	-0.046689868	0.9425	-0.041927855	0.9735	0.8103
3.1485	88500	0.0003	-	-	-	-	-	-	-	-
3.1662	89000	0.0003	-	-	-	-	-	-	-	-
3.1840	89500	0.0003	-	-	-	-	-	-	-	-
3.2018	90000	0.0003	0.0003	-0.048173092	0.9318	-0.046601273	0.9422	-0.041765593	0.9741	0.8067
3.2196	90500	0.0003	-	-	-	-	-	-	-	-
3.2374	91000	0.0003	-	-	-	-	-	-	-	-
3.2552	91500	0.0003	-	-	-	-	-	-	-	-
3.2730	92000	0.0003	0.0003	-0.04806846	0.9333	-0.046525463	0.9425	-0.041717805	0.9735	0.8069
3.2908	92500	0.0003	-	-	-	-	-	-	-	-
3.3086	93000	0.0003	-	-	-	-	-	-	-	-
3.3263	93500	0.0003	-	-	-	-	-	-	-	-
3.3441	94000	0.0003	0.0003	-0.04804208	0.9337	-0.04647612	0.9432	-0.041626267	0.9737	0.8102
3.3619	94500	0.0003	-	-	-	-	-	-	-	-
3.3797	95000	0.0003	-	-	-	-	-	-	-	-
3.3975	95500	0.0003	-	-	-	-	-	-	-	-
3.4153	96000	0.0003	0.0003	-0.047956012	0.9349	-0.046421595	0.9436	-0.04157625	0.9745	0.8101
3.4331	96500	0.0003	-	-	-	-	-	-	-	-
3.4509	97000	0.0003	-	-	-	-	-	-	-	-
3.4686	97500	0.0003	-	-	-	-	-	-	-	-
3.4864	98000	0.0003	0.0003	-0.04786362	0.9340	-0.046320286	0.9440	-0.04152808	0.9747	0.8075
3.5042	98500	0.0003	-	-	-	-	-	-	-	-
3.5220	99000	0.0003	-	-	-	-	-	-	-	-
3.5398	99500	0.0003	-	-	-	-	-	-	-	-
3.5576	100000	0.0003	0.0003	-0.04788445	0.9331	-0.04631946	0.9435	-0.041442778	0.9738	0.8050
3.5754	100500	0.0003	-	-	-	-	-	-	-	-
3.5932	101000	0.0003	-	-	-	-	-	-	-	-
3.6110	101500	0.0003	-	-	-	-	-	-	-	-
3.6287	102000	0.0003	0.0003	-0.047823686	0.9340	-0.046267267	0.9434	-0.041403364	0.9738	0.8120
3.6465	102500	0.0003	-	-	-	-	-	-	-	-
3.6643	103000	0.0003	-	-	-	-	-	-	-	-
3.6821	103500	0.0003	-	-	-	-	-	-	-	-
3.6999	104000	0.0003	0.0003	-0.047792524	0.9339	-0.046203755	0.9437	-0.04136616	0.9744	0.8145
3.7177	104500	0.0003	-	-	-	-	-	-	-	-
3.7355	105000	0.0003	-	-	-	-	-	-	-	-
3.7533	105500	0.0003	-	-	-	-	-	-	-	-
3.7710	106000	0.0003	0.0003	-0.047719087	0.9336	-0.046151668	0.9434	-0.04131788	0.9739	0.8092
3.7888	106500	0.0003	-	-	-	-	-	-	-	-
3.8066	107000	0.0003	-	-	-	-	-	-	-	-
3.8244	107500	0.0003	-	-	-	-	-	-	-	-
3.8422	108000	0.0003	0.0003	-0.047663696	0.9340	-0.046100874	0.9437	-0.041290164	0.9743	0.8110
3.8600	108500	0.0003	-	-	-	-	-	-	-	-
3.8778	109000	0.0003	-	-	-	-	-	-	-	-
3.8956	109500	0.0003	-	-	-	-	-	-	-	-
3.9134	110000	0.0003	0.0003	-0.04758864	0.9342	-0.045986332	0.9441	-0.04112862	0.9746	0.8107
3.9311	110500	0.0003	-	-	-	-	-	-	-	-
3.9489	111000	0.0003	-	-	-	-	-	-	-	-
3.9667	111500	0.0003	-	-	-	-	-	-	-	-
3.9845	112000	0.0003	0.0003	-0.047555514	0.9334	-0.04596298	0.9442	-0.041138608	0.9747	0.8179
4.0023	112500	0.0003	-	-	-	-	-	-	-	-
4.0201	113000	0.0003	-	-	-	-	-	-	-	-
4.0379	113500	0.0003	-	-	-	-	-	-	-	-
4.0556	114000	0.0003	0.0003	-0.047533713	0.9337	-0.045918535	0.9436	-0.041128766	0.9738	0.8093
4.0734	114500	0.0003	-	-	-	-	-	-	-	-
4.0912	115000	0.0003	-	-	-	-	-	-	-	-
4.1090	115500	0.0003	-	-	-	-	-	-	-	-
4.1268	116000	0.0003	0.0003	-0.047445226	0.9337	-0.04587704	0.9437	-0.04101509	0.9745	0.8124
4.1446	116500	0.0003	-	-	-	-	-	-	-	-
4.1624	117000	0.0003	-	-	-	-	-	-	-	-
4.1802	117500	0.0003	-	-	-	-	-	-	-	-
4.1979	118000	0.0003	0.0003	-0.047467772	0.9333	-0.04589099	0.944	-0.04104045	0.9747	0.8118
4.2157	118500	0.0003	-	-	-	-	-	-	-	-
4.2335	119000	0.0003	-	-	-	-	-	-	-	-
4.2513	119500	0.0003	-	-	-	-	-	-	-	-
4.2691	120000	0.0003	0.0003	-0.047381546	0.9332	-0.045793895	0.944	-0.040915057	0.974	0.8102
4.2869	120500	0.0003	-	-	-	-	-	-	-	-
4.3047	121000	0.0003	-	-	-	-	-	-	-	-
4.3225	121500	0.0003	-	-	-	-	-	-	-	-
4.3403	122000	0.0003	0.0003	-0.047399823	0.9331	-0.045784745	0.9444	-0.040900033	0.9744	0.8168
4.3580	122500	0.0003	-	-	-	-	-	-	-	-
4.3758	123000	0.0003	-	-	-	-	-	-	-	-
4.3936	123500	0.0003	-	-	-	-	-	-	-	-
4.4114	124000	0.0003	0.0003	-0.04735861	0.9335	-0.045788392	0.9442	-0.04084509	0.9746	0.8194
4.4292	124500	0.0003	-	-	-	-	-	-	-	-
4.4470	125000	0.0003	-	-	-	-	-	-	-	-
4.4648	125500	0.0003	-	-	-	-	-	-	-	-
4.4826	126000	0.0003	0.0003	-0.04727602	0.9330	-0.04571327	0.9435	-0.04079213	0.9747	0.8191
4.5003	126500	0.0003	-	-	-	-	-	-	-	-
4.5181	127000	0.0003	-	-	-	-	-	-	-	-
4.5359	127500	0.0003	-	-	-	-	-	-	-	-
4.5537	128000	0.0003	0.0003	-0.04724487	0.9328	-0.045612063	0.9444	-0.040754147	0.9748	0.8109
4.5715	128500	0.0003	-	-	-	-	-	-	-	-
4.5893	129000	0.0003	-	-	-	-	-	-	-	-
4.6071	129500	0.0003	-	-	-	-	-	-	-	-
4.6249	130000	0.0003	0.0003	-0.047237676	0.9351	-0.04558688	0.9440	-0.040685352	0.9747	0.8180
4.6427	130500	0.0003	-	-	-	-	-	-	-	-
4.6604	131000	0.0003	-	-	-	-	-	-	-	-
4.6782	131500	0.0003	-	-	-	-	-	-	-	-
4.6960	132000	0.0003	0.0003	-0.047139905	0.9355	-0.04554107	0.9441	-0.040673897	0.9751	0.8108
4.7138	132500	0.0003	-	-	-	-	-	-	-	-
4.7316	133000	0.0003	-	-	-	-	-	-	-	-
4.7494	133500	0.0003	-	-	-	-	-	-	-	-
4.7672	134000	0.0003	0.0003	-0.047172163	0.9341	-0.04553711	0.9432	-0.04068736	0.9751	0.8122
4.7850	134500	0.0003	-	-	-	-	-	-	-	-
4.8027	135000	0.0003	-	-	-	-	-	-	-	-
4.8205	135500	0.0003	-	-	-	-	-	-	-	-
4.8383	136000	0.0003	0.0003	-0.04712533	0.9348	-0.045525618	0.9437	-0.040563405	0.9747	0.8150
4.8561	136500	0.0003	-	-	-	-	-	-	-	-
4.8739	137000	0.0003	-	-	-	-	-	-	-	-
4.8917	137500	0.0003	-	-	-	-	-	-	-	-
4.9095	138000	0.0003	0.0003	-0.047049224	0.9345	-0.045493715	0.9433	-0.040560503	0.9744	0.8134
4.9273	138500	0.0003	-	-	-	-	-	-	-	-
4.9451	139000	0.0003	-	-	-	-	-	-	-	-
4.9628	139500	0.0003	-	-	-	-	-	-	-	-
4.9806	140000	0.0003	0.0003	-0.047017988	0.9340	-0.045397308	0.944	-0.040537592	0.9745	0.8143
4.9984	140500	0.0003	-	-	-	-	-	-	-	-
5.0162	141000	0.0003	-	-	-	-	-	-	-	-
5.0340	141500	0.0003	-	-	-	-	-	-	-	-
5.0518	142000	0.0003	0.0003	-0.047043595	0.9355	-0.045408167	0.9443	-0.040538292	0.9748	0.8144
5.0696	142500	0.0003	-	-	-	-	-	-	-	-
5.0873	143000	0.0003	-	-	-	-	-	-	-	-
5.1051	143500	0.0003	-	-	-	-	-	-	-	-
5.1229	144000	0.0003	0.0003	-0.04705367	0.9355	-0.045425817	0.9442	-0.040479798	0.9747	0.8121
5.1407	144500	0.0003	-	-	-	-	-	-	-	-
5.1585	145000	0.0003	-	-	-	-	-	-	-	-
5.1763	145500	0.0003	-	-	-	-	-	-	-	-
5.1941	146000	0.0003	0.0003	-0.046963915	0.9352	-0.04530141	0.9442	-0.04043464	0.9747	0.8117
5.2119	146500	0.0003	-	-	-	-	-	-	-	-
5.2296	147000	0.0003	-	-	-	-	-	-	-	-
5.2474	147500	0.0003	-	-	-	-	-	-	-	-
5.2652	148000	0.0003	0.0003	-0.04690311	0.9348	-0.045300145	0.9454	-0.040396217	0.9757	0.8158
5.2830	148500	0.0003	-	-	-	-	-	-	-	-
5.3008	149000	0.0003	-	-	-	-	-	-	-	-
5.3186	149500	0.0003	-	-	-	-	-	-	-	-
5.3364	150000	0.0003	0.0003	-0.04689872	0.9344	-0.04525921	0.9444	-0.040379316	0.9751	0.8148
5.3542	150500	0.0003	-	-	-	-	-	-	-	-
5.3720	151000	0.0003	-	-	-	-	-	-	-	-
5.3897	151500	0.0003	-	-	-	-	-	-	-	-
5.4075	152000	0.0003	0.0003	-0.046854757	0.9349	-0.04521157	0.9442	-0.040353984	0.9747	0.8115
5.4253	152500	0.0003	-	-	-	-	-	-	-	-
5.4431	153000	0.0003	-	-	-	-	-	-	-	-
5.4609	153500	0.0003	-	-	-	-	-	-	-	-
5.4787	154000	0.0003	0.0003	-0.046820126	0.9353	-0.045219637	0.9447	-0.04032354	0.975	0.8158
5.4965	154500	0.0003	-	-	-	-	-	-	-	-
5.5143	155000	0.0003	-	-	-	-	-	-	-	-
5.5320	155500	0.0003	-	-	-	-	-	-	-	-
5.5498	156000	0.0003	0.0003	-0.046734486	0.9351	-0.045157243	0.9443	-0.04024718	0.9748	0.8164
5.5676	156500	0.0003	-	-	-	-	-	-	-	-
5.5854	157000	0.0003	-	-	-	-	-	-	-	-
5.6032	157500	0.0003	-	-	-	-	-	-	-	-
5.6210	158000	0.0003	0.0003	-0.046766065	0.9356	-0.045193	0.9448	-0.04020705	0.9750	0.8161
5.6388	158500	0.0003	-	-	-	-	-	-	-	-
5.6566	159000	0.0003	-	-	-	-	-	-	-	-
5.6744	159500	0.0003	-	-	-	-	-	-	-	-
5.6921	160000	0.0003	0.0003	-0.046653003	0.9349	-0.045091264	0.9447	-0.040132213	0.9752	0.8159
5.7099	160500	0.0003	-	-	-	-	-	-	-	-
5.7277	161000	0.0003	-	-	-	-	-	-	-	-
5.7455	161500	0.0003	-	-	-	-	-	-	-	-
5.7633	162000	0.0003	0.0003	-0.046689577	0.9350	-0.045103475	0.9439	-0.040139068	0.9752	0.8148
5.7811	162500	0.0003	-	-	-	-	-	-	-	-
5.7989	163000	0.0003	-	-	-	-	-	-	-	-
5.8167	163500	0.0003	-	-	-	-	-	-	-	-
5.8344	164000	0.0003	0.0003	-0.04665651	0.9354	-0.04505335	0.9442	-0.040089488	0.9745	0.8172
5.8522	164500	0.0003	-	-	-	-	-	-	-	-
5.8700	165000	0.0003	-	-	-	-	-	-	-	-
5.8878	165500	0.0003	-	-	-	-	-	-	-	-
5.9056	166000	0.0003	0.0003	-0.046620764	0.9353	-0.045048986	0.9445	-0.04012648	0.9752	0.8171
5.9234	166500	0.0003	-	-	-	-	-	-	-	-
5.9412	167000	0.0003	-	-	-	-	-	-	-	-
5.9590	167500	0.0003	-	-	-	-	-	-	-	-
5.9768	168000	0.0003	0.0003	-0.046547405	0.9346	-0.044982277	0.9449	-0.03998514	0.9749	0.8148
5.9945	168500	0.0003	-	-	-	-	-	-	-	-
6.0123	169000	0.0003	-	-	-	-	-	-	-	-
6.0301	169500	0.0003	-	-	-	-	-	-	-	-
6.0479	170000	0.0003	0.0003	-0.046575766	0.9353	-0.044996176	0.9445	-0.040046997	0.9748	0.8173
6.0657	170500	0.0003	-	-	-	-	-	-	-	-
6.0835	171000	0.0003	-	-	-	-	-	-	-	-
6.1013	171500	0.0003	-	-	-	-	-	-	-	-
6.1190	172000	0.0003	0.0003	-0.04653697	0.9344	-0.04489271	0.9442	-0.039989993	0.9751	0.8117
6.1368	172500	0.0003	-	-	-	-	-	-	-	-
6.1546	173000	0.0003	-	-	-	-	-	-	-	-
6.1724	173500	0.0003	-	-	-	-	-	-	-	-
6.1902	174000	0.0003	0.0003	-0.046540864	0.9354	-0.044921882	0.9445	-0.039948173	0.9753	0.8154
6.2080	174500	0.0003	-	-	-	-	-	-	-	-
6.2258	175000	0.0003	-	-	-	-	-	-	-	-
6.2436	175500	0.0003	-	-	-	-	-	-	-	-
6.2613	176000	0.0003	0.0003	-0.046473563	0.935	-0.044842277	0.9445	-0.039841603	0.9749	0.8173
6.2791	176500	0.0003	-	-	-	-	-	-	-	-
6.2969	177000	0.0003	-	-	-	-	-	-	-	-
6.3147	177500	0.0003	-	-	-	-	-	-	-	-
6.3325	178000	0.0003	0.0003	-0.046434972	0.9357	-0.044857066	0.9446	-0.03987878	0.9747	0.8137
6.3503	178500	0.0003	-	-	-	-	-	-	-	-
6.3681	179000	0.0003	-	-	-	-	-	-	-	-
6.3859	179500	0.0003	-	-	-	-	-	-	-	-
6.4037	180000	0.0003	0.0003	-0.046446227	0.9348	-0.044805177	0.9443	-0.03994019	0.9748	0.8147
6.4214	180500	0.0003	-	-	-	-	-	-	-	-
6.4392	181000	0.0003	-	-	-	-	-	-	-	-
6.4570	181500	0.0003	-	-	-	-	-	-	-	-
6.4748	182000	0.0003	0.0003	-0.046411477	0.9350	-0.044835974	0.9449	-0.039863113	0.9753	0.8177
6.4926	182500	0.0003	-	-	-	-	-	-	-	-
6.5104	183000	0.0003	-	-	-	-	-	-	-	-
6.5282	183500	0.0003	-	-	-	-	-	-	-	-
6.5460	184000	0.0003	0.0003	-0.04641683	0.9354	-0.04482221	0.9449	-0.039854627	0.9752	0.8135
6.5637	184500	0.0003	-	-	-	-	-	-	-	-
6.5815	185000	0.0003	-	-	-	-	-	-	-	-
6.5993	185500	0.0003	-	-	-	-	-	-	-	-
6.6171	186000	0.0003	0.0003	-0.04635241	0.9343	-0.044773895	0.9448	-0.039753385	0.9750	0.8183
6.6349	186500	0.0003	-	-	-	-	-	-	-	-
6.6527	187000	0.0003	-	-	-	-	-	-	-	-
6.6705	187500	0.0003	-	-	-	-	-	-	-	-
6.6883	188000	0.0003	0.0003	-0.046331976	0.9351	-0.044707794	0.9445	-0.039761487	0.9752	0.8197
6.7060	188500	0.0003	-	-	-	-	-	-	-	-
6.7238	189000	0.0003	-	-	-	-	-	-	-	-
6.7416	189500	0.0003	-	-	-	-	-	-	-	-
6.7594	190000	0.0003	0.0003	-0.046344783	0.9355	-0.04470515	0.9448	-0.03973746	0.9758	0.8171

Framework Versions

• Python: 3.10.16
• Sentence Transformers: 3.3.1
• Transformers: 4.48.0
• PyTorch: 2.5.1+cu124
• Accelerate: 1.2.1
• Datasets: 3.2.0
• 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",
}