sartifyllc/MultiLinguSwahili-nomic-embed-text-v1.5-nli-matryoshka

SentenceTransformer based on nomic-ai/nomic-embed-text-v1.5

This is a sentence-transformers model finetuned from nomic-ai/nomic-embed-text-v1.5 on the Mollel/swahili-n_li-triplet-swh-eng dataset. 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.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: nomic-ai/nomic-embed-text-v1.5
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • Mollel/swahili-n_li-triplet-swh-eng

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': 8192, 'do_lower_case': False}) with Transformer model: NomicBertModel 
  (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("sartifyllc/MultiLinguSwahili-nomic-embed-text-v1.5-nli-matryoshka")
# Run inference
sentences = [
    'Mwanamume na mwanamke wachanga waliovaa mikoba wanaweka au kuondoa kitu kutoka kwenye mti mweupe wa zamani, huku watu wengine wamesimama au wameketi nyuma.',
    'mwanamume na mwanamke wenye mikoba',
    'tai huruka',
]
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]

Evaluation

Metrics

Semantic Similarity

• Dataset: sts-test-768
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6945
spearman_cosine	0.6872
pearson_manhattan	0.7086
spearman_manhattan	0.7136
pearson_euclidean	0.7084
spearman_euclidean	0.7128
pearson_dot	0.4819
spearman_dot	0.4659
pearson_max	0.7086
spearman_max	0.7136

Semantic Similarity

• Dataset: sts-test-512
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6926
spearman_cosine	0.6859
pearson_manhattan	0.7087
spearman_manhattan	0.7128
pearson_euclidean	0.7089
spearman_euclidean	0.7124
pearson_dot	0.4684
spearman_dot	0.4526
pearson_max	0.7089
spearman_max	0.7128

Semantic Similarity

• Dataset: sts-test-256
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6877
spearman_cosine	0.6815
pearson_manhattan	0.7084
spearman_manhattan	0.7098
pearson_euclidean	0.7094
spearman_euclidean	0.7104
pearson_dot	0.4439
spearman_dot	0.4255
pearson_max	0.7094
spearman_max	0.7104

Semantic Similarity

• Dataset: sts-test-128
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6709
spearman_cosine	0.667
pearson_manhattan	0.7042
spearman_manhattan	0.7001
pearson_euclidean	0.7055
spearman_euclidean	0.7023
pearson_dot	0.3786
spearman_dot	0.3593
pearson_max	0.7055
spearman_max	0.7023

Semantic Similarity

• Dataset: sts-test-64
• Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.6534
spearman_cosine	0.6524
pearson_manhattan	0.692
spearman_manhattan	0.6857
pearson_euclidean	0.695
spearman_euclidean	0.6899
pearson_dot	0.335
spearman_dot	0.3097
pearson_max	0.695
spearman_max	0.6899

Training Details

Training Dataset

Mollel/swahili-n_li-triplet-swh-eng

• Dataset: Mollel/swahili-n_li-triplet-swh-eng
• Size: 1,115,700 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: 15.18 tokens max: 80 tokens	min: 5 tokens mean: 18.53 tokens max: 52 tokens	min: 5 tokens mean: 17.8 tokens max: 53 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.
  Mtu aliyepanda farasi anaruka juu ya ndege iliyovunjika.	Mtu yuko nje, juu ya farasi.	Mtu yuko kwenye mkahawa, akiagiza omelette.
  Children smiling and waving at camera	There are children present	The kids are frowning

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Evaluation Dataset

Mollel/swahili-n_li-triplet-swh-eng

• Dataset: Mollel/swahili-n_li-triplet-swh-eng
• Size: 13,168 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: 26.43 tokens max: 94 tokens	min: 5 tokens mean: 13.37 tokens max: 65 tokens	min: 5 tokens mean: 14.7 tokens max: 54 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.
  Wanawake wawili wanakumbatiana huku wakishikilia vifurushi vya kwenda.	Wanawake wawili wanashikilia vifurushi.	Wanaume hao wanapigana nje ya duka la vyakula vitamu.
  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.

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 24
• per_device_eval_batch_size: 24
• learning_rate: 2e-05
• 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
• prediction_loss_only: True
• per_device_train_batch_size: 24
• per_device_eval_batch_size: 24
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• learning_rate: 2e-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
• 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, '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_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Click to expand

Epoch	Step	Training Loss	sts-test-128_spearman_cosine	sts-test-256_spearman_cosine	sts-test-512_spearman_cosine	sts-test-64_spearman_cosine	sts-test-768_spearman_cosine
0.0043	100	10.0627	-	-	-	-	-
0.0086	200	8.2355	-	-	-	-	-
0.0129	300	6.7233	-	-	-	-	-
0.0172	400	6.5832	-	-	-	-	-
0.0215	500	6.7512	-	-	-	-	-
0.0258	600	6.7634	-	-	-	-	-
0.0301	700	6.5592	-	-	-	-	-
0.0344	800	5.0689	-	-	-	-	-
0.0387	900	4.7079	-	-	-	-	-
0.0430	1000	4.6359	-	-	-	-	-
0.0473	1100	4.4513	-	-	-	-	-
0.0516	1200	4.2328	-	-	-	-	-
0.0559	1300	3.7454	-	-	-	-	-
0.0602	1400	3.9198	-	-	-	-	-
0.0645	1500	4.0727	-	-	-	-	-
0.0688	1600	3.8923	-	-	-	-	-
0.0731	1700	3.8137	-	-	-	-	-
0.0774	1800	4.1512	-	-	-	-	-
0.0817	1900	4.1304	-	-	-	-	-
0.0860	2000	4.0195	-	-	-	-	-
0.0903	2100	3.6836	-	-	-	-	-
0.0946	2200	2.9968	-	-	-	-	-
0.0990	2300	2.8909	-	-	-	-	-
0.1033	2400	3.0884	-	-	-	-	-
0.1076	2500	3.3081	-	-	-	-	-
0.1119	2600	3.6266	-	-	-	-	-
0.1162	2700	4.3754	-	-	-	-	-
0.1205	2800	4.0218	-	-	-	-	-
0.1248	2900	3.7167	-	-	-	-	-
0.1291	3000	3.4815	-	-	-	-	-
0.1334	3100	3.6446	-	-	-	-	-
0.1377	3200	3.44	-	-	-	-	-
0.1420	3300	3.6725	-	-	-	-	-
0.1463	3400	3.4699	-	-	-	-	-
0.1506	3500	3.076	-	-	-	-	-
0.1549	3600	3.1179	-	-	-	-	-
0.1592	3700	3.1704	-	-	-	-	-
0.1635	3800	3.4614	-	-	-	-	-
0.1678	3900	4.1157	-	-	-	-	-
0.1721	4000	4.1584	-	-	-	-	-
0.1764	4100	4.5602	-	-	-	-	-
0.1807	4200	3.6875	-	-	-	-	-
0.1850	4300	4.1521	-	-	-	-	-
0.1893	4400	3.5475	-	-	-	-	-
0.1936	4500	3.4036	-	-	-	-	-
0.1979	4600	3.0564	-	-	-	-	-
0.2022	4700	3.7761	-	-	-	-	-
0.2065	4800	3.6857	-	-	-	-	-
0.2108	4900	3.3534	-	-	-	-	-
0.2151	5000	4.1137	-	-	-	-	-
0.2194	5100	3.5239	-	-	-	-	-
0.2237	5200	4.1297	-	-	-	-	-
0.2280	5300	3.5339	-	-	-	-	-
0.2323	5400	3.9294	-	-	-	-	-
0.2366	5500	3.717	-	-	-	-	-
0.2409	5600	3.3346	-	-	-	-	-
0.2452	5700	4.0495	-	-	-	-	-
0.2495	5800	3.7869	-	-	-	-	-
0.2538	5900	3.9533	-	-	-	-	-
0.2581	6000	4.1135	-	-	-	-	-
0.2624	6100	3.6655	-	-	-	-	-
0.2667	6200	3.9111	-	-	-	-	-
0.2710	6300	3.8582	-	-	-	-	-
0.2753	6400	3.7712	-	-	-	-	-
0.2796	6500	3.6536	-	-	-	-	-
0.2839	6600	3.4516	-	-	-	-	-
0.2882	6700	3.7151	-	-	-	-	-
0.2925	6800	3.7659	-	-	-	-	-
0.2969	6900	3.3159	-	-	-	-	-
0.3012	7000	3.5753	-	-	-	-	-
0.3055	7100	4.2095	-	-	-	-	-
0.3098	7200	3.718	-	-	-	-	-
0.3141	7300	4.0709	-	-	-	-	-
0.3184	7400	3.8079	-	-	-	-	-
0.3227	7500	3.3735	-	-	-	-	-
0.3270	7600	3.7303	-	-	-	-	-
0.3313	7700	3.2693	-	-	-	-	-
0.3356	7800	3.6564	-	-	-	-	-
0.3399	7900	3.6702	-	-	-	-	-
0.3442	8000	3.7274	-	-	-	-	-
0.3485	8100	3.8536	-	-	-	-	-
0.3528	8200	3.9516	-	-	-	-	-
0.3571	8300	3.7351	-	-	-	-	-
0.3614	8400	3.649	-	-	-	-	-
0.3657	8500	3.5913	-	-	-	-	-
0.3700	8600	3.7733	-	-	-	-	-
0.3743	8700	3.6359	-	-	-	-	-
0.3786	8800	4.2983	-	-	-	-	-
0.3829	8900	3.6692	-	-	-	-	-
0.3872	9000	3.7309	-	-	-	-	-
0.3915	9100	3.8886	-	-	-	-	-
0.3958	9200	3.8999	-	-	-	-	-
0.4001	9300	3.5528	-	-	-	-	-
0.4044	9400	3.6309	-	-	-	-	-
0.4087	9500	4.2475	-	-	-	-	-
0.4130	9600	3.793	-	-	-	-	-
0.4173	9700	3.6575	-	-	-	-	-
0.4216	9800	3.84	-	-	-	-	-
0.4259	9900	3.3721	-	-	-	-	-
0.4302	10000	4.3743	-	-	-	-	-
0.4345	10100	3.5054	-	-	-	-	-
0.4388	10200	3.54	-	-	-	-	-
0.4431	10300	3.6197	-	-	-	-	-
0.4474	10400	3.7567	-	-	-	-	-
0.4517	10500	3.9814	-	-	-	-	-
0.4560	10600	3.6277	-	-	-	-	-
0.4603	10700	3.5071	-	-	-	-	-
0.4646	10800	3.8348	-	-	-	-	-
0.4689	10900	3.8674	-	-	-	-	-
0.4732	11000	3.0325	-	-	-	-	-
0.4775	11100	3.7262	-	-	-	-	-
0.4818	11200	3.6921	-	-	-	-	-
0.4861	11300	3.4946	-	-	-	-	-
0.4904	11400	3.7541	-	-	-	-	-
0.4948	11500	3.6751	-	-	-	-	-
0.4991	11600	3.8765	-	-	-	-	-
0.5034	11700	3.5058	-	-	-	-	-
0.5077	11800	3.5135	-	-	-	-	-
0.5120	11900	3.8052	-	-	-	-	-
0.5163	12000	3.3015	-	-	-	-	-
0.5206	12100	3.5389	-	-	-	-	-
0.5249	12200	3.5226	-	-	-	-	-
0.5292	12300	3.6715	-	-	-	-	-
0.5335	12400	3.2256	-	-	-	-	-
0.5378	12500	3.3447	-	-	-	-	-
0.5421	12600	3.6315	-	-	-	-	-
0.5464	12700	3.8674	-	-	-	-	-
0.5507	12800	3.4066	-	-	-	-	-
0.5550	12900	3.7356	-	-	-	-	-
0.5593	13000	3.5742	-	-	-	-	-
0.5636	13100	3.7676	-	-	-	-	-
0.5679	13200	3.7907	-	-	-	-	-
0.5722	13300	3.8089	-	-	-	-	-
0.5765	13400	3.4742	-	-	-	-	-
0.5808	13500	3.6536	-	-	-	-	-
0.5851	13600	3.7736	-	-	-	-	-
0.5894	13700	3.9072	-	-	-	-	-
0.5937	13800	3.7386	-	-	-	-	-
0.5980	13900	3.3387	-	-	-	-	-
0.6023	14000	3.5509	-	-	-	-	-
0.6066	14100	3.7056	-	-	-	-	-
0.6109	14200	3.7283	-	-	-	-	-
0.6152	14300	3.7301	-	-	-	-	-
0.6195	14400	3.8027	-	-	-	-	-
0.6238	14500	3.5606	-	-	-	-	-
0.6281	14600	3.9467	-	-	-	-	-
0.6324	14700	3.3394	-	-	-	-	-
0.6367	14800	4.1254	-	-	-	-	-
0.6410	14900	3.7121	-	-	-	-	-
0.6453	15000	3.9167	-	-	-	-	-
0.6496	15100	3.8084	-	-	-	-	-
0.6539	15200	3.7794	-	-	-	-	-
0.6582	15300	3.7664	-	-	-	-	-
0.6625	15400	3.4378	-	-	-	-	-
0.6668	15500	3.6632	-	-	-	-	-
0.6711	15600	3.8493	-	-	-	-	-
0.6754	15700	4.1475	-	-	-	-	-
0.6797	15800	3.5782	-	-	-	-	-
0.6840	15900	3.4341	-	-	-	-	-
0.6883	16000	3.3295	-	-	-	-	-
0.6927	16100	3.8165	-	-	-	-	-
0.6970	16200	3.9702	-	-	-	-	-
0.7013	16300	3.6555	-	-	-	-	-
0.7056	16400	3.6946	-	-	-	-	-
0.7099	16500	3.8027	-	-	-	-	-
0.7142	16600	3.4523	-	-	-	-	-
0.7185	16700	3.461	-	-	-	-	-
0.7228	16800	3.4403	-	-	-	-	-
0.7271	16900	3.6398	-	-	-	-	-
0.7314	17000	3.8443	-	-	-	-	-
0.7357	17100	3.6012	-	-	-	-	-
0.7400	17200	3.6645	-	-	-	-	-
0.7443	17300	3.4899	-	-	-	-	-
0.7486	17400	3.7186	-	-	-	-	-
0.7529	17500	3.6199	-	-	-	-	-
0.7572	17600	4.4274	-	-	-	-	-
0.7615	17700	4.0262	-	-	-	-	-
0.7658	17800	3.9325	-	-	-	-	-
0.7701	17900	3.6338	-	-	-	-	-
0.7744	18000	3.6136	-	-	-	-	-
0.7787	18100	3.4514	-	-	-	-	-
0.7830	18200	3.4427	-	-	-	-	-
0.7873	18300	3.3601	-	-	-	-	-
0.7916	18400	3.313	-	-	-	-	-
0.7959	18500	3.4062	-	-	-	-	-
0.8002	18600	3.098	-	-	-	-	-
0.8045	18700	3.183	-	-	-	-	-
0.8088	18800	3.1482	-	-	-	-	-
0.8131	18900	3.0122	-	-	-	-	-
0.8174	19000	3.0828	-	-	-	-	-
0.8217	19100	3.063	-	-	-	-	-
0.8260	19200	2.9688	-	-	-	-	-
0.8303	19300	3.0425	-	-	-	-	-
0.8346	19400	3.2018	-	-	-	-	-
0.8389	19500	2.9111	-	-	-	-	-
0.8432	19600	2.9516	-	-	-	-	-
0.8475	19700	2.9115	-	-	-	-	-
0.8518	19800	2.9323	-	-	-	-	-
0.8561	19900	2.8753	-	-	-	-	-
0.8604	20000	2.8344	-	-	-	-	-
0.8647	20100	2.7665	-	-	-	-	-
0.8690	20200	2.7732	-	-	-	-	-
0.8733	20300	2.8622	-	-	-	-	-
0.8776	20400	2.8749	-	-	-	-	-
0.8819	20500	2.8534	-	-	-	-	-
0.8863	20600	2.9254	-	-	-	-	-
0.8906	20700	2.7366	-	-	-	-	-
0.8949	20800	2.7287	-	-	-	-	-
0.8992	20900	2.9469	-	-	-	-	-
0.9035	21000	2.9052	-	-	-	-	-
0.9078	21100	2.7256	-	-	-	-	-
0.9121	21200	2.8469	-	-	-	-	-
0.9164	21300	2.6626	-	-	-	-	-
0.9207	21400	2.6796	-	-	-	-	-
0.9250	21500	2.6927	-	-	-	-	-
0.9293	21600	2.7125	-	-	-	-	-
0.9336	21700	2.6734	-	-	-	-	-
0.9379	21800	2.7199	-	-	-	-	-
0.9422	21900	2.6635	-	-	-	-	-
0.9465	22000	2.5218	-	-	-	-	-
0.9508	22100	2.7595	-	-	-	-	-
0.9551	22200	2.6821	-	-	-	-	-
0.9594	22300	2.6578	-	-	-	-	-
0.9637	22400	2.568	-	-	-	-	-
0.9680	22500	2.5527	-	-	-	-	-
0.9723	22600	2.6857	-	-	-	-	-
0.9766	22700	2.6637	-	-	-	-	-
0.9809	22800	2.6311	-	-	-	-	-
0.9852	22900	2.4635	-	-	-	-	-
0.9895	23000	2.6239	-	-	-	-	-
0.9938	23100	2.6873	-	-	-	-	-
0.9981	23200	2.5138	-	-	-	-	-
1.0	23244	-	0.6670	0.6815	0.6859	0.6524	0.6872

Framework Versions

• Python: 3.11.9
• Sentence Transformers: 3.0.1
• Transformers: 4.40.1
• PyTorch: 2.3.0+cu121
• Accelerate: 0.29.3
• Datasets: 2.19.0
• 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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}