mics-nlp/xlm-roberta-small-all-nli-triplet

SentenceTransformer based on Unbabel/xlm-roberta-comet-small

This is a sentence-transformers model finetuned from Unbabel/xlm-roberta-comet-small on the sentence-transformers/all-nli dataset. 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: Unbabel/xlm-roberta-comet-small
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 384 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • sentence-transformers/all-nli
• Language: en

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': 384, '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("mics-nlp/xlm-roberta-small-all-nli-triplet")
# Run inference
sentences = [
    'a baby smiling',
    'A baby is unhappy.',
    'The dog has big ears.',
]
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

Triplet

• Dataset: all-nli-dev
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.849
dot_accuracy	0.163
manhattan_accuracy	0.837
euclidean_accuracy	0.841
max_accuracy	0.849

Triplet

• Dataset: all-nli-test
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.839
dot_accuracy	0.15
manhattan_accuracy	0.827
euclidean_accuracy	0.827
max_accuracy	0.839

Training Details

Training Dataset

sentence-transformers/all-nli

• Dataset: sentence-transformers/all-nli at d482672
• Size: 100,000 training samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 7 tokens mean: 10.9 tokens max: 52 tokens	min: 6 tokens mean: 13.62 tokens max: 42 tokens	min: 5 tokens mean: 14.76 tokens max: 55 tokens

• Samples:

  anchor	positive	negative
  A person on a horse jumps over a broken down airplane.	A person is outdoors, on a horse.	A person is at a diner, ordering an omelette.
  Children smiling and waving at camera	There are children present	The kids are frowning
  A boy is jumping on skateboard in the middle of a red bridge.	The boy does a skateboarding trick.	The boy skates down the sidewalk.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

sentence-transformers/all-nli

• Dataset: sentence-transformers/all-nli at d482672
• Size: 1,000 evaluation samples
• Columns: anchor, positive, and negative
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative
  type	string	string	string
  details	min: 6 tokens mean: 20.31 tokens max: 83 tokens	min: 5 tokens mean: 10.71 tokens max: 35 tokens	min: 5 tokens mean: 11.39 tokens max: 32 tokens

• Samples:

  anchor	positive	negative
  Two women are embracing while holding to go packages.	Two woman are holding packages.	The men are fighting outside a deli.
  Two young children in blue jerseys, one with the number 9 and one with the number 2 are standing on wooden steps in a bathroom and washing their hands in a sink.	Two kids in numbered jerseys wash their hands.	Two kids in jackets walk to school.
  A man selling donuts to a customer during a world exhibition event held in the city of Angeles	A man selling donuts to a customer.	A woman drinks her coffee in a small cafe.

• Loss: CachedMultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 1
• warmup_ratio: 0.1
• bf16: True
• batch_sampler: no_duplicates

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
• 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: 1
• 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: 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: 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: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• 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
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	loss	all-nli-dev_max_accuracy	all-nli-test_max_accuracy
0	0	-	-	0.541	-
0.016	100	3.5308	3.1817	0.558	-
0.032	200	3.2784	3.0406	0.597	-
0.048	300	3.113	2.7572	0.635	-
0.064	400	2.8296	2.4646	0.68	-
0.08	500	2.631	2.3583	0.676	-
0.096	600	2.3247	2.1394	0.706	-
0.112	700	2.2211	2.0201	0.711	-
0.128	800	2.1263	1.9560	0.757	-
0.144	900	2.2105	1.9074	0.748	-
0.16	1000	2.0637	1.9289	0.728	-
0.176	1100	2.1772	1.8796	0.741	-
0.192	1200	2.1518	1.8346	0.761	-
0.208	1300	1.728	1.8213	0.765	-
0.224	1400	1.8101	1.6321	0.772	-
0.24	1500	1.7516	1.5669	0.793	-
0.256	1600	1.4988	1.5538	0.8	-
0.272	1700	1.6695	1.5462	0.803	-
0.288	1800	1.5971	1.5499	0.783	-
0.304	1900	1.5614	1.5047	0.788	-
0.32	2000	1.522	1.4957	0.794	-
0.336	2100	1.3624	1.4153	0.814	-
0.352	2200	1.4773	1.4169	0.809	-
0.368	2300	1.6066	1.3697	0.813	-
0.384	2400	1.5106	1.3203	0.819	-
0.4	2500	1.4783	1.3417	0.817	-
0.416	2600	1.3696	1.2650	0.824	-
0.432	2700	1.5115	1.2779	0.829	-
0.448	2800	1.4834	1.2668	0.834	-
0.464	2900	1.4823	1.2621	0.836	-
0.48	3000	1.4163	1.2465	0.837	-
0.496	3100	1.4232	1.2475	0.837	-
0.512	3200	1.2193	1.1975	0.838	-
0.528	3300	1.2569	1.1816	0.838	-
0.544	3400	1.2988	1.1936	0.839	-
0.56	3500	1.5068	1.2213	0.835	-
0.576	3600	1.3022	1.1799	0.842	-
0.592	3700	1.3823	1.1910	0.831	-
0.608	3800	1.4224	1.1786	0.834	-
0.624	3900	1.3765	1.1541	0.843	-
0.64	4000	1.4987	1.1365	0.844	-
0.656	4100	1.7525	1.1394	0.843	-
0.672	4200	1.6013	1.1178	0.841	-
0.688	4300	1.3326	1.0959	0.846	-
0.704	4400	1.355	1.0757	0.848	-
0.72	4500	1.2834	1.0681	0.846	-
0.736	4600	1.2939	1.0696	0.85	-
0.752	4700	1.4069	1.0645	0.848	-
0.768	4800	1.4503	1.0609	0.849	-
0.784	4900	1.2833	1.0587	0.847	-
0.8	5000	1.3321	1.0563	0.849	-
0.816	5100	1.3006	1.0539	0.847	-
0.832	5200	1.4332	1.0527	0.847	-
0.848	5300	1.3101	1.0505	0.848	-
0.864	5400	1.3658	1.0523	0.849	-
0.88	5500	1.353	1.0520	0.849	-
0.896	5600	1.2429	1.0521	0.848	-
0.912	5700	1.3512	1.0505	0.848	-
0.928	5800	1.2995	1.0501	0.848	-
0.944	5900	1.3514	1.0491	0.849	-
0.96	6000	1.3976	1.0490	0.848	-
0.976	6100	1.2112	1.0487	0.848	-
0.992	6200	0.0033	1.0492	0.849	-
1.0	6250	-	-	-	0.839

Framework Versions

• Python: 3.9.10
• Sentence Transformers: 3.0.0
• Transformers: 4.41.2
• PyTorch: 2.3.0+cu121
• Accelerate: 0.26.1
• Datasets: 2.16.1
• Tokenizers: 0.19.1

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",
}

CachedMultipleNegativesRankingLoss

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup}, 
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}