metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - dense
  - generated_from_trainer
  - dataset_size:800
  - loss:MultipleNegativesRankingLoss
base_model: microsoft/mpnet-base
widget:
  - source_sentence: What is the department of medicine located at?
    sentences:
      - >-
        Publisher’s Note: MDPI stays neutral

        with regard to jurisdictional claims in

        published maps and institutional afil-


        iations.


        onon)


        Copyright: © 2021 by the author.

        Licensee MDPI, Basel, Switzerland.

        This article is an open access article

        distributed under the terms and

        conditions of the Creative Commons

        Attribution (CC BY) license (https://

        creativecommons.org/licenses/by/

        4.0/).


        Joan and Sanford I. Weill Department of Medicine, Weill Cornell Medical
        College, 525 East 68th Street,

        Room M-522, Box 130, New York, NY 10065, USA; [email protected] or
        [email protected]
      - >-
        Results At the parameters used, the ultrasound did not directly affect
        bCSC proliferation, with no evident changes in

        morphology. In contrast, the ultrasound protocol affected the migration
        and invasion ability of bCSCs, limiting their

        capacity to advance while a major affection was detected on their
        angiogenic properties. LIPUS-treated bCSCs were

        unable to transform into aggressive metastatic cancer cells, by
        decreasing their migration and invasion capacity as

        well as vessel formation. Finally, RNA-seq analysis revealed major
        changes in gene expression, with 676 differentially
      - |-
        Tesfaye, M. & Savoldo, B. Adoptive cell therapy in
        treating pediatric solid tumors. Curr. Oncol. Rep. 20,
        73 (2018).

        Marofi, F. et al. CAR T cells in solid tumors: challenges
        and opportunities. Stem Cell Res. Ther. 12, 81 (2021).
        Deng, Q. et al. Characteristics of anti-CD19 CAR T cell
        infusion products associated with efficacy and toxicity

        in patients with large B cell lymphomas. Nat. Med. 26,

        1878-1887 (2020).
        Boulch, M. A cross-talk between CAR T cell subsets
        and the tumor microenvironment is essential for
        sustained cytotoxic activity. Sci. Immunol. 6,
        eabd4344 (2021).
  - source_sentence: >-
      What is the result of LIPUS treatment on the formation of new vessels and
      tubes?
    sentences:
      - >-
        apparatus), and mitochondrial damage, which then leads to eventual cell
        death [112,114].

        Accordingly, alterations that affect the lysosomal-mitochondria
        relationship and their

        metabolic equilibrium generate a defective metabolism, which contributes
        to disease pro-

        gression [115]. Consequently, the identification of regulatory molecular
        links between these

        two organelles will most probably cause the rise of novel targets for
        the treatment of NPC.

        Therefore, we propose that members of the miRNA-17-92 cluster could be
        relevant actors
      - |-
        A tube formation assay was conducted on Matrigel to
        study the impact of LIPUS stimulation on bCSCs’ angio-
        genic activity (Fig. 5). After 2 h, both control and LIPUS-
        stimulated cells exhibited signs of angiogenesis (Fig. 5A
        and B). This observation was further confirmed by count-
        ing the number of panel-like structures and vessels in
        both conditions, which were slightly higher in control
        cells (Fig. 5C). Statistical analysis using Student’s t-test
        revealed that LIPUS treatment significantly reduced the
        formation of new vessels and tubes (y=0.0039). These
      - |-
        Although a number of preclinical studies, like the ones
        previously described, have shown considerable promise re-
        garding the use of ADSC-therapy, more studies are needed.
        Future studies can continue to work toward determining if
        hADSCs are capable of being used for cell replacement and
        better elucidate the mechanisms by which hADSCs work.

        IV. ADIPOSE TISSUE AS A SOURCE FOR STEM
        CELLS
  - source_sentence: What percentage of cases had malignant lesions?
    sentences:
      - >-
        Vedolizumab Monoclonal antibody anti «487 integrins, blocks gut homing
        of T lymphocytes


        “These drugs are used as second line treatments for SR aGvHD, as
        reviewed by Penack et al. (11).

        ’Ruxolitinib has been recently approved by FDA as second line therapy
        for SR aGVHD.


        TABLE 3 | Major drugs used as second line treatment of cGvHD and their
        mechanisms.


        Drug* Major mechanisms identified


        Cyclosporin A, tacrolimus Calcineurin inhibitors that block downstrem
        TCR signalling leading to NFAT regulated genes transcription; block T
        cells

        activation
      - >-
        --- Page 4 ---

        J. Clin. Med. 2024, 13, 7559


        4 of 13


        lesions were found in 59 cases (70.24%) and malignant lesions in 25
        cases (29.76%). In DC

        IV, benign lesions were found in 57 cases (81.4%) and malignant lesions
        in 13 cases (18.6%).

        There were no statistically significant associations between gender (p =
        0.76), BMI (p = 0.52),

        and obesity (p = 0.76) and the presence of thyroid malignancy.


        Table 1. Demographic and pathologic features of 521 patients who
        underwent surgery due to


        thyroid nodules.
      - |-
        MSCs showed that these exosomes induce angiogenesis in
        endothelial cells via the activation of the NF«B pathway (141).
        However, in another study exosomes derived from hypoxia-
        preconditioned MSCs contributed to the attenuation of the
        injury resulting from an ischemia/reperfusion episode via the
        Wnt signaling pathway (142). Beyond that, hypoxia seems to
        increase exosome secretion in general (141). Also, in a fat
        graft model, co-transplantation of exosomes from hypoxia pre-
        conditioned adipose-derived MSC improved vascularization and
        graft survival (143) (see Table 5).
  - source_sentence: >-
      When is routine fine-needle aspiration biopsy (FS) recommended during
      thyroidectomy?
    sentences:
      - >-
        ing queries about its routine use due to the improved preoperative
        diagnosis. Nowadays, while the use of FS during thyroidectomy

        has decreased, it is still used as an additional method for different
        purposes intraoperatively. FS may not always provide definitive

        results. If FS will alter the surgical plan or extent, it should be
        applied. Routine FS is not recommended for evaluating thyroid nod-

        ules. But in addition to FNAB, if FS results may change the operation
        plan or extent, they can be utilized. FS should not be applied
      - |-
        Approximately 15% of FNABs take part in this category.
        After their initial Bethesda | FNAB, the malignancy risk in
        nodules surgically excised, ranges between 5-20%. Repeat
        FNAB is recommended if the initial FNAB result is Bethes-
        da |, and in 60-80% of cases, the repeat FNAB results in a
        diagnostic category.''?*°! If the second FNAB also yields a
        nondiagnostic result, surgical resection is recommended.
        21] Especially in cases with Bethesda | FNAB and with a sur-
        gical indication, an intraoperative FS can be utilized.® It
        has been reported that FS significantly contributes to the
      - |-
        Preconditioning with a myriad of other soluble factors, such
        as growth factors or hormones, seems to also potentiate MSCs
        regenerative capacity, mainly by stimulating angiogenesis and
        inhibiting fibrosis. For example, intracardiac transplantation
        of SDF-1-preconditioned MSCs increased angiogenesis and
        reduced fibrosis in the ischemic area of a post-infarct heart (89).
        The effects observed were attributed to the activation of the Akt
        signaling pathway, similarly to what was described for hypoxia-
        preconditioned MSCs. TGF-a-preconditioned MSCs enhanced
  - source_sentence: >-
      What is the number of genes obtained from comparing control and
      LIPUS-stimulated samples?
    sentences:
      - |-
        Differentially expressed genes (DEGs) were obtained
        between control and LIPUS-stimulated samples using
        an adjusted P<0.05 and|log2FC| > 1 as cutoffs to define
        statistically significant differential expression. 676 genes
        were obtained from which 578 were upregulated when
        stimulated with LIPUS and 98 genes were subregulated
        (Supp. Figure 1). To further understand the functions
        and pathways associated with the differentially expressed
        genes (DEG), Gene Ontology (GO) and Kyoto Encyclo-
        pedia of Genes and Genomes (KEGG) analyses were con-
        ducted using the DAVID database [37, 38].
      - |-
        Another advantage of ADSCs is their immune privilege
        status due to a lack of major histocompatibility complex
        II (MHC Il) and costimulatory molecules.42,43,45,.47 Some
        studies have even demonstrated a higher immunosuppres-
        sion capacity in ADSCs compared to BMSCs as ADSCs ex-
        pressed lower levels of human antigen class I (HLA I) anti-
        gen.47 They also have a unique secretome and can produce
        immunomodulatory, anti-apoptotic, hematopoietic, and
        angiogenic factors that can help with repair of tissues -
        characteristics that may support successful transplanta-
      - >-
        independent studies have shown a raising trend in both cancer incidence
        [2] and a high-salt

        dietary lifestyle [7], there is no direct correlation between dietary
        salt intake and breast

        cancer. Interestingly, in the human body, certain organs such as the
        skin and lymph nodes

        have a natural tendency to accumulate salt [8]. Although unknown, the
        pathophysiological

        significance of this selective accumulation of sodium in certain organs
        and solid tumors is

        an area of intense research.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - cosine_accuracy
model-index:
  - name: SentenceTransformer based on microsoft/mpnet-base
    results:
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: initial test
          type: initial_test
        metrics:
          - type: cosine_accuracy
            value: 0.9800000190734863
            name: Cosine Accuracy
      - task:
          type: triplet
          name: Triplet
        dataset:
          name: final test
          type: final_test
        metrics:
          - type: cosine_accuracy
            value: 0.9800000190734863
            name: Cosine Accuracy

SentenceTransformer based on microsoft/mpnet-base

This is a sentence-transformers model finetuned from microsoft/mpnet-base on the json 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: microsoft/mpnet-base
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity
Training Dataset:
- json

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, 'architecture': 'MPNetModel'})
  (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("sahithkumar7/final-mpnet-base-fullfinetuned-epoch3")
# Run inference
sentences = [
    'What is the number of genes obtained from comparing control and LIPUS-stimulated samples?',
    'Differentially expressed genes (DEGs) were obtained\nbetween control and LIPUS-stimulated samples using\nan adjusted P<0.05 and|log2FC| > 1 as cutoffs to define\nstatistically significant differential expression. 676 genes\nwere obtained from which 578 were upregulated when\nstimulated with LIPUS and 98 genes were subregulated\n(Supp. Figure 1). To further understand the functions\nand pathways associated with the differentially expressed\ngenes (DEG), Gene Ontology (GO) and Kyoto Encyclo-\npedia of Genes and Genomes (KEGG) analyses were con-\nducted using the DAVID database [37, 38].',
    'independent studies have shown a raising trend in both cancer incidence [2] and a high-salt\ndietary lifestyle [7], there is no direct correlation between dietary salt intake and breast\ncancer. Interestingly, in the human body, certain organs such as the skin and lymph nodes\nhave a natural tendency to accumulate salt [8]. Although unknown, the pathophysiological\nsignificance of this selective accumulation of sodium in certain organs and solid tumors is\nan area of intense research.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000,  0.6291, -0.0130],
#         [ 0.6291,  1.0000, -0.0026],
#         [-0.0130, -0.0026,  1.0000]])

Evaluation

Metrics

Triplet

Datasets: initial_test and final_test
Evaluated with TripletEvaluator

Metric	initial_test	final_test
cosine_accuracy	0.98	0.98

Training Details

Training Dataset

json

Dataset: json
Size: 800 training samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 800 samples:

	anchor	positive	negative
type	string	string	string
details	min: 7 tokens mean: 16.79 tokens max: 39 tokens	min: 39 tokens mean: 117.74 tokens max: 265 tokens	min: 40 tokens mean: 116.14 tokens max: 356 tokens

Samples:

anchor	positive	negative
`What is the limitation of FBG-based sensors in tactile feedback?`	Furthermore, FBG-based 3-axis tactile sensors have been proposed for a more comprehensive haptic perception tool in surgeries (Figure 1D) (16). Five optical fibers merged with FBG sensors are suspended in a deformable medium and measure the compression or tension of the tissue as the sensors are pressed against it, returning a _ surface reaction map. While FBG-based sensors are small, flexible, and sensitive, there are several challenges that need to be addressed for optimal performance for tactile feedback. These sensors are temperature sensitive, requiring temperature	141]. Therefore, it is not known to what extent spared axons are remyelinated by transplanted Schwann cells, nor is the contribution of this myelin to functional im- provements proven. Transplantation of Schwann cells incapable of producing myelin, such as cells derived from trembler (Pmp22Tr) mutant mice, may be useful in establishing a causal relationship between myelin re- generation and functional improvements. Several MSC transplantations demonstrate an increase of myelin re- tention and the number of myelinated axons in the le- sion site during a chronic post-injury period [57]. Thus,
`What are the advantages of strain elastography?`	frontiersin.org --- Page 8 --- Kumar et al. TABLE 2 Modalities of ultrasound elastography. Modality Strain elastography Excitation Applied manual compression (38) Advantages No additional specialized equipment required (40) 10.3389/fmedt.2023.1238129 Limitations Qualitative measurements (39) Internal physiological mechanism (42) Simple low-cost design (40) Applied compression is operator-dependent (51) More commonly used (52) High inter-observer variability (51) coustic radiation force impulse Acoustic radiation force (43) (ARFI) imaging Image beyond slip boundaries (45)	Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afil- iations. onon) Copyright: © 2021 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/). Joan and Sanford I. Weill Department of Medicine, Weill Cornell Medical College, 525 East 68th Street, Room M-522, Box 130, New York, NY 10065, USA; [email protected] or [email protected]
`What is the material used for the substrate in a piezoelectric element?`	`gain for biomedical applications. frontiersin.org --- Page 9 --- Kumar et al. > [PMUT ] Electrode: Voltage Electrode2 © piezoelectric elements o —: OSi02 ©) silicon substrate B [ CMUT ] AC DC membrane ————— vacuum insulator substrate = ground FIGURE 3`	Histopatholo Cytology Total, n (%) Benign, n (%) P ey Cancer, n (%) FA 2 (15.4%) FTC 2 (25%) 0 GD (7.7%) PTC 6 (75%) I 21 (4.0%) NG 9 (69.2%) Other diagnosis (7.7%) FA 15 (9.9%) FIC 4 (14.3%) FT-UMP (0.7%) MTC 3 (10.7%) GD (0.7%) PTC 21 (75%) Il 180 (34.5%) OA (0.7%) LT (0.7%) NG 130 (85.5%) NIFTP 2 (1.3%) FA 14 (23.7%) FIC 7 (28.0%) FI-UMP 2 (3.4%) OTC 1 (4.0%) OA (1.7%) PTC 17 (68.0%) Il 84 (16.1%) LT 3 (5.1%) NG 35 (59.3%) NIFTP 2 (3.4%) WDT-UMP 2 (3.4%) FA 15 (26.3%) OTC 1 (7.7%) FT-UMP 5 (8.8%) PTC 12 (92.3%) OA 13 (22.8%) IV 70 (13.4%) LT 2 (3.5%) NG 18 (31.6%) NIFTP 2 (3.5%)

Loss: MultipleNegativesRankingLoss with these parameters:

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

Evaluation Dataset

json

Dataset: json
Size: 200 evaluation samples
Columns: anchor, positive, and negative

Approximate statistics based on the first 200 samples:

	anchor	positive	negative
type	string	string	string
details	min: 7 tokens mean: 17.14 tokens max: 35 tokens	min: 40 tokens mean: 121.3 tokens max: 356 tokens	min: 45 tokens mean: 119.75 tokens max: 356 tokens

Samples:

anchor	positive	negative
`What can differentiate into a very wide variety of tissues?`	lead to decreased rates of graft-versus-host disease. They also can differentiate into a very wide variety of tissues. For example, when compared with bone marrow stem cells or mobilized peripheral blood, umbilical cord blood stem cells have a greater repopulating ability.5° Cord blood derived CD34+ cells have very potent hematopoietic abilities, and this is attributed to the immaturity of the stem cells rela- tive to adult derived cells. Studies have been done that an- alyze long term survival of children with hematologic dis- orders who were transplanted with umbilical cord blood	metabolic regulation may affect the function of more than one organelle. Therefore, if the miR-17-92 regulatory cluster can perturb genes related to mitochondrial metabolic function, it could be also related, in some way, to genes involved in lysosomal metabolic function. Lysosomes are intracellular organelles that, in form of small vesicles, participate in several cellular functions, mainly digestion, but also vesicle trafficking, autophagy, nutrient sensing, cellular growth, signaling [85], and even enzyme secretion. The membrane-bound
`What are the two most common types of pluripotent stem cells?`	III]. AMNIOTIC CELLS AS A SOURCE FOR STEM CELLS Historically, the two most common types of pluripotent stem cells include embryonic stem cells (ESCs) and induced pluripotent stem cells (iPSCs).35 However, despite the many research efforts to improve ESC and iPSC technologies, there are still enormous clinical challenges.°> Two signif- icant issues posed by ESC and iPSC technologies include low survival rate of transplanted cells and tumorigenicity.°> Recently, researchers have isolated pluripotent stem cells	`Explanation: criterion 6 indicates a positive diagnosis only within the DC VI group relative to all other categories. Criterion 5 indicates a positive diagnosis within the DCs VI and V relative to all other categories. The highest positive predictive value (PPV) confirming malignancy through histopatho- logical examination for criterion 6 was 0.93, and for criterion 5, it was 0.92. For the subsequent criteria, the PPVs were as follows: criterion 4—0.66; criterion 3—0.55; criterion 2—0.40.`
`What percentage of stem cells are present in bone marrow?`	`ing 30% in some tissues.43-45 This is a significant difference from the .0001-.0002% stem cells present in bone marrow.43 Given this difference in stem cell concentration between the sources, there will be more ADSCs per sample of WAT`	migration of bCSCs. This finding raises the possibil- ity that LIPUS may decrease the ability of these cells to invade adjacent tissues and start the process of metasta- ses. These results also suggested that some of the changes induced by LIPUS take longer to be detected in this type of 2D migration model, possible due to changes in gene expression pattern. To further study this hypothesis, we performed a Transwell invasion assay. The data revealed a reduced number of cells crossing the membrane after LIPUS stimulation, indicating that therapeutic LIPUS

Loss: MultipleNegativesRankingLoss 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
warmup_ratio: 0.1
fp16: 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
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: 3
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: True
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: 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: no_duplicates
multi_dataset_batch_sampler: proportional
router_mapping: {}
learning_rate_mapping: {}

Training Logs

Click to expand

Epoch	Step	Training Loss	Validation Loss	initial_test_cosine_accuracy	final_test_cosine_accuracy
-1	-1	-	-	0.7800	-
0.02	1	3.124	-	-	-
0.04	2	3.2227	-	-	-
0.06	3	3.1108	-	-	-
0.08	4	3.1317	-	-	-
0.1	5	3.3326	-	-	-
0.12	6	2.9797	-	-	-
0.14	7	3.0933	-	-	-
0.16	8	2.7409	-	-	-
0.18	9	2.7381	-	-	-
0.2	10	2.6301	-	-	-
0.22	11	2.005	-	-	-
0.24	12	2.1863	-	-	-
0.26	13	2.8065	-	-	-
0.28	14	1.6524	-	-	-
0.3	15	1.7121	-	-	-
0.32	16	1.9863	-	-	-
0.34	17	1.4783	-	-	-
0.36	18	1.0542	-	-	-
0.38	19	1.1223	-	-	-
0.4	20	1.0425	0.9097	0.9000	-
0.42	21	1.2517	-	-	-
0.44	22	1.048	-	-	-
0.46	23	1.0064	-	-	-
0.48	24	0.9887	-	-	-
0.5	25	0.6468	-	-	-
0.52	26	0.8978	-	-	-
0.54	27	0.439	-	-	-
0.56	28	0.8051	-	-	-
0.58	29	0.7684	-	-	-
0.6	30	0.573	-	-	-
0.62	31	0.6101	-	-	-
0.64	32	0.9438	-	-	-
0.66	33	0.8656	-	-	-
0.68	34	0.5758	-	-	-
0.7	35	0.2412	-	-	-
0.72	36	0.4738	-	-	-
0.74	37	0.7844	-	-	-
0.76	38	0.7517	-	-	-
0.78	39	0.3222	-	-	-
0.8	40	0.466	0.6199	0.9600	-
0.82	41	0.5259	-	-	-
0.84	42	0.3936	-	-	-
0.86	43	0.23	-	-	-
0.88	44	0.4184	-	-	-
0.9	45	0.7641	-	-	-
0.92	46	0.2579	-	-	-
0.94	47	1.2493	-	-	-
0.96	48	0.4205	-	-	-
0.98	49	0.4778	-	-	-
1.0	50	0.545	-	-	-
1.02	51	0.2018	-	-	-
1.04	52	0.2048	-	-	-
1.06	53	0.2031	-	-	-
1.08	54	0.5784	-	-	-
1.1	55	0.2764	-	-	-
1.12	56	0.5112	-	-	-
1.1400	57	0.2482	-	-	-
1.16	58	0.3772	-	-	-
1.18	59	0.1247	-	-	-
1.2	60	0.1832	0.5882	1.0	-
1.22	61	0.1802	-	-	-
1.24	62	0.3174	-	-	-
1.26	63	0.0836	-	-	-
1.28	64	0.2814	-	-	-
1.3	65	0.0926	-	-	-
1.32	66	0.3834	-	-	-
1.34	67	0.2547	-	-	-
1.3600	68	0.3229	-	-	-
1.38	69	0.0441	-	-	-
1.4	70	0.1735	-	-	-
1.42	71	0.0494	-	-	-
1.44	72	0.2197	-	-	-
1.46	73	0.2218	-	-	-
1.48	74	0.2196	-	-	-
1.5	75	0.2516	-	-	-
1.52	76	0.6337	-	-	-
1.54	77	0.1729	-	-	-
1.56	78	0.5629	-	-	-
1.58	79	0.4224	-	-	-
1.6	80	0.1977	0.4683	1.0	-
1.62	81	0.2117	-	-	-
1.6400	82	0.2423	-	-	-
1.6600	83	0.2047	-	-	-
1.6800	84	0.1741	-	-	-
1.7	85	0.4539	-	-	-
1.72	86	0.5744	-	-	-
1.74	87	0.2697	-	-	-
1.76	88	0.1926	-	-	-
1.78	89	0.1882	-	-	-
1.8	90	0.1527	-	-	-
1.8200	91	0.2154	-	-	-
1.8400	92	0.5145	-	-	-
1.8600	93	0.1294	-	-	-
1.88	94	0.1499	-	-	-
1.9	95	0.2143	-	-	-
1.92	96	0.2039	-	-	-
1.94	97	0.1662	-	-	-
1.96	98	0.1414	-	-	-
1.98	99	0.0743	-	-	-
2.0	100	0.1603	0.4067	0.9800	-
2.02	101	0.1885	-	-	-
2.04	102	0.1539	-	-	-
2.06	103	0.0592	-	-	-
2.08	104	0.0874	-	-	-
2.1	105	0.0873	-	-	-
2.12	106	0.057	-	-	-
2.14	107	0.0317	-	-	-
2.16	108	0.0807	-	-	-
2.18	109	0.0232	-	-	-
2.2	110	0.0847	-	-	-
2.22	111	0.0811	-	-	-
2.24	112	0.0688	-	-	-
2.26	113	0.1392	-	-	-
2.2800	114	0.0681	-	-	-
2.3	115	0.0329	-	-	-
2.32	116	0.0177	-	-	-
2.34	117	0.0794	-	-	-
2.36	118	0.1128	-	-	-
2.38	119	0.095	-	-	-
2.4	120	0.0384	0.4131	0.9800	-
2.42	121	0.0791	-	-	-
2.44	122	0.078	-	-	-
2.46	123	0.0232	-	-	-
2.48	124	0.0265	-	-	-
2.5	125	0.023	-	-	-
2.52	126	0.1105	-	-	-
2.54	127	0.0114	-	-	-
2.56	128	0.1051	-	-	-
2.58	129	0.0178	-	-	-
2.6	130	0.0731	-	-	-
2.62	131	0.051	-	-	-
2.64	132	0.0589	-	-	-
2.66	133	0.1714	-	-	-
2.68	134	0.0452	-	-	-
2.7	135	0.0491	-	-	-
2.7200	136	0.0652	-	-	-
2.74	137	0.0534	-	-	-
2.76	138	0.0414	-	-	-
2.7800	139	0.0611	-	-	-
2.8	140	0.1983	0.4193	0.9800	-
2.82	141	0.0489	-	-	-
2.84	142	0.0215	-	-	-
2.86	143	0.0491	-	-	-
2.88	144	0.0521	-	-	-
2.9	145	0.1212	-	-	-
2.92	146	0.0464	-	-	-
2.94	147	0.0145	-	-	-
2.96	148	0.0281	-	-	-
2.98	149	0.1358	-	-	-
3.0	150	0.0479	-	-	-
-1	-1	-	-	-	0.9800

Framework Versions

Python: 3.11.13
Sentence Transformers: 5.0.0
Transformers: 4.52.4
PyTorch: 2.6.0+cu124
Accelerate: 1.8.1
Datasets: 3.6.0
Tokenizers: 0.21.2

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",
}

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}
}