Add new SentenceTransformer model

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

1_Pooling/config.json

@@ -0,0 +1,10 @@
+{
+    "word_embedding_dimension": 1024,
+    "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
+}

README.md

@@ -0,0 +1,928 @@
+---
+tags:
+- sentence-transformers
+- sentence-similarity
+- feature-extraction
+- dense
+- generated_from_trainer
+- dataset_size:17319
+- loss:MultipleNegativesRankingLoss
+base_model: intfloat/multilingual-e5-large
+widget:
+- source_sentence: 'query: The unexpected repercussions of COVID-19 vaccine policy:
+    why requirements, certificates and limitations could do more damage than benefit
+    | BMJ Global Health'
+  sentences:
+  - 'passage: title: SARS-CoV-2 infects and replicates in cells of the human endocrine
+    and exocrine pancreas abstract: Infection-related diabetes can arise as a result
+    of virus-associated β-cell destruction.
+
+    Clinical data suggest that the severe acute respiratory syndrome coronavirus 2
+    (SARS-CoV-2), causing the coronavirus disease 2019 (COVID-19), impairs glucose
+    homoeostasis, but experimental evidence that SARS-CoV-2 can infect pancreatic
+    tissue has been lacking.
+
+    In the present study, we show that SARS-CoV-2 infects cells of the human exocrine
+    and endocrine pancreas ex vivo and in vivo.
+
+    We demonstrate that human β-cells express viral entry proteins, and SARS-CoV-2
+    infects and replicates in cultured human islets.
+
+    Infection is associated with morphological, transcriptional and functional changes,
+    including reduced numbers of insulin-secretory granules in β-cells and impaired
+    glucose-stimulated insulin secretion.
+
+    In COVID-19 full-body postmortem examinations, we detected SARS-CoV-2 nucleocapsid
+    protein in pancreatic exocrine cells, and in cells that stain positive for the
+    β-cell marker NKX6.
+
+    1 and are in close proximity to the islets of Langerhans in all four patients
+    investigated.
+
+    Our data identify the human pancreas as a target of SARS-CoV-2 infection and suggest
+    that β-cell infection could contribute to the metabolic dysregulation observed
+    in patients with COVID-19.
+
+    SARS-CoV-2 is shown to infect and replicate in human pancreatic tissue, including
+    in β-cells, which is associated with morphological, transcriptomic and functional
+    changes.'
+  - 'passage: title: A SARS-like cluster of circulating bat coronaviruses shows potential
+    for human emergence abstract: The emergence of severe acute respiratory syndrome
+    coronavirus (SARS-CoV) and Middle East respiratory syndrome (MERS)-CoV underscores
+    the threat of cross-species transmission events leading to outbreaks in humans.
+
+    Here we examine the disease potential of a SARS-like virus, SHC014-CoV, which
+    is currently circulating in Chinese horseshoe bat populations.
+
+    Using the SARS-CoV reverse genetics system, we generated and characterized a chimeric
+    virus expressing the spike of bat coronavirus SHC014 in a mouse-adapted SARS-CoV
+    backbone.
+
+    The results indicate that group 2b viruses encoding the SHC014 spike in a wild-type
+    backbone can efficiently use multiple orthologs of the SARS receptor human angiotensin
+    converting enzyme II (ACE2), replicate efficiently in primary human airway cells
+    and achieve in vitro titers equivalent to epidemic strains of SARS-CoV.
+
+    Additionally, in vivo experiments demonstrate replication of the chimeric virus
+    in mouse lung with notable pathogenesis.
+
+    Evaluation of available SARS-based immune-therapeutic and prophylactic modalities
+    revealed poor efficacy; both monoclonal antibody and vaccine approaches failed
+    to neutralize and protect from infection with CoVs using the novel spike protein.
+
+    On the basis of these findings, we synthetically re-derived an infectious full-length
+    SHC014 recombinant virus and demonstrate robust viral replication both in vitro
+    and in vivo.
+
+    Our work suggests a potential risk of SARS-CoV re-emergence from viruses currently
+    circulating in bat populations.'
+  - 'passage: title: The unintended consequences of COVID-19 vaccine policy: why mandates,
+    passports and restrictions may cause more harm than good abstract: Vaccination
+    policies have shifted dramatically during COVID-19 with the rapid emergence of
+    population-wide vaccine mandates, domestic vaccine passports and differential
+    restrictions based on vaccination status.
+
+    While these policies have prompted ethical, scientific, practical, legal and political
+    debate, there has been limited evaluation of their potential unintended consequences.
+
+    Here, we outline a comprehensive set of hypotheses for why these policies may
+    ultimately be counterproductive and harmful.
+
+    Our framework considers four domains: (1) behavioural psychology, (2) politics
+    and law, (3) socioeconomics, and (4) the integrity of science and public health.
+
+    While current vaccines appear to have had a significant impact on decreasing COVID-19-related
+    morbidity and mortality burdens, we argue that current mandatory vaccine policies
+    are scientifically questionable and are likely to cause more societal harm than
+    good.
+
+    Restricting people’s access to work, education, public transport and social life
+    based on COVID-19 vaccination status impinges on human rights, promotes stigma
+    and social polarisation, and adversely affects health and well-being.
+
+    Current policies may lead to a widening of health and economic inequalities, detrimental
+    long-term impacts on trust in government and scientific institutions, and reduce
+    the uptake of future public health measures, including COVID-19 vaccines as well
+    as routine immunisations.
+
+    Mandating vaccination is one of the most powerful interventions in public health
+    and should be used sparingly and carefully to uphold ethical norms and trust in
+    institutions.'
+- source_sentence: 'query: @VanessaBeeley No vaccine ! Results indicate following
+    standard case of light COVID-19, SARSCOV2 CD8+ Tcells not only persist but steadily
+    differentiate in a synchronized manner deep into convalescence into a state characteristic
+    of long-lived, self-renewing memory.'
+  sentences:
+  - 'passage: title: Protracted yet Coordinated Differentiation of Long-Lived SARS-CoV-2-Specific
+    CD8+ T Cells during Convalescence abstract: Abstract CD8+ T cells can potentiate
+    long-lived immunity against COVID-19.
+
+    We screened longitudinally-sampled convalescent human donors against SARS-CoV-2
+    tetramers and identified a participant with an immunodominant response against
+    residues 322 to 311 of nucleocapsid (Nuc322–331), a peptide conserved in all variants
+    of concern reported to date.
+
+    We conducted 38-parameter cytometry by time of flight on tetramer-identified Nuc322–331–specific
+    CD8+ T cells and on CD4+ and CD8+ T cells recognizing the entire nucleocapsid
+    and spike proteins, and took 32 serological measurements.
+
+    We discovered a coordination of the Nuc322–331–specific CD8+ T response with both
+    the CD4+ T cell and Ab pillars of adaptive immunity.
+
+    Over the approximately six month period of convalescence monitored, we observed
+    a slow and progressive decrease in the activation state and polyfunctionality
+    of Nuc322–331–specific CD8+ T cells, accompanied by an increase in their lymph
+    node–homing and homeostatic proliferation potential.
+
+    These results suggest that following a typical case of mild COVID-19, SARS-CoV-2–specific
+    CD8+ T cells not only persist but continuously differentiate in a coordinated
+    fashion well into convalescence into a state characteristic of long-lived, self-renewing
+    memory.'
+  - 'passage: publication has also been changed; recommendations are now ordered by
+    severity of covid-19.
+
+
+    About this guideline This living guideline from the World Health Organization
+    (WHO) incorporates new evidence to dynamically update recommendations for covid-19
+    therapeutics.
+
+    The GDG typically evaluates a therapy when the WHO judges sufficient evidence
+    is available to make a recommendation.
+
+    While the GDG takes an individual patient perspective in making recommendations,
+    it also considers resource implications, acceptability, feasibility, equity, and
+    human rights.
+
+    This guideline was developed according to standards and methods for trustworthy
+    guidelines, making use of an innovative process to achieve efficiency in dynamic
+    updating of recommendations.
+
+    The methods are aligned with the WHO Handbook for Guideline Development and according
+    to a pre-approved protocol (planning proposal) by the Guideline Review Committee
+    (GRC).
+
+    A box at the end of the article outlines key methodological aspects of the guideline
+    process.
+
+    MAGIC Evidence Ecosystem Foundation provides methodological support, including
+    the coordination of living systematic reviews with network meta-analyses to inform
+    the recommendations.
+
+    The full version of the guideline is available online in MAGICapp and in PDF on
+    the WHO website, with a summary version here in The BMJ .
+
+    These formats should facilitate adaptation, which is strongly encouraged by WHO
+    to contextualise recommendations in a healthcare system to maximise impact.
+
+    Future recommendations Recommendations on anticoagulation are planned for the
+    next update to this guideline.
+
+    Updated data regarding systemic corticosteroids, azithromycin, favipiravir and
+    umefenovir for non-severe illness, and convalescent plasma and statin therapy
+    for severe or critical illness, are planned for review in upcoming guideline iterations.'
+  - 'passage: title: Understanding the Effects of Age and T-Cell Differentiation on
+    COVID-19 Severity: Implicating a Fas/FasL-mediated Feed-Forward Controller of
+    T-Cell Differentiation abstract: OPINION article Front.
+
+    Immunol.
+
+    , 03 March 2022Sec.
+
+    T Cell Biology Volume 13 - 2022 | https://doi.
+
+    org/10.
+
+    3389/fimmu.
+
+    2022.'
+- source_sentence: 'query: Preprint:Beschwerden nach Covid-Impfung🧵 "Von den 1 704
+    teilnehmenden Beschäftigten des Gesundheitswesens waren insgesamt 595 (34,9 %)
+    nach mindestens einer COVID‑19‑Impfung krankgeschrieben, was zu einer Gesamtzahl
+    von 1 550 Krankheitstagen führte."'
+  sentences:
+  - 'passage: title: Projecting the transmission dynamics of SARS-CoV-2 through the
+    postpandemic period abstract: It is urgent to understand the future of severe
+    acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) transmission.
+
+    We used estimates of seasonality, immunity, and cross-immunity for human coronavirus
+    OC43 (HCoV-OC43) and HCoV-HKU1 using time-series data from the United States to
+    inform a model of SARS-CoV-2 transmission.
+
+    We projected that recurrent wintertime outbreaks of SARS-CoV-2 will probably occur
+    after the initial, most severe pandemic wave.
+
+    Absent other interventions, a key metric for the success of social distancing
+    is whether critical care capacities are exceeded.
+
+    To avoid this, prolonged or intermittent social distancing may be necessary into
+    2022.
+
+    Additional interventions, including expanded critical care capacity and an effective
+    therapeutic, would improve the success of intermittent distancing and hasten the
+    acquisition of herd immunity.
+
+    Longitudinal serological studies are urgently needed to determine the extent and
+    duration of immunity to SARS-CoV-2.
+
+    Even in the event of apparent elimination, SARS-CoV-2 surveillance should be maintained
+    because a resurgence in contagion could be possible as late as 2024.'
+  - 'passage: title: Inability to work following COVID-19 vaccination among healthcare
+    workers - an important aspect for future booster vaccinations abstract: ABSTRACT
+    Background COVID-19 vaccination is a key prevention strategy to reduce the spread
+    and severity of SARS-CoV-2 infections, especially among highly exposed healthcare
+    workers (HCWs).
+
+    However, vaccine-related inability to work among HCWs could overstrain healthcare
+    systems.
+
+    Methods This study examined sick leave and intake of pro re nata (PRN) medication
+    after the first, second and third COVID-19 vaccination in HCWs.
+
+    Subgroup analyses were performed for different vaccines, gender, healthcare professions,
+    and for HCWs aged at least 30 years.
+
+    Data was collected by using an electronic questionnaire.
+
+    Findings Among 1,704 HCWs enrolled, in total 595 (34·9%) HCWs were on sick leave
+    following at least one COVID-19 vaccination, leading to a total number of 1,550
+    sick days.
+
+    Both the absolute sick days and the rate of HCWs on sick leave significantly increased
+    with each subsequent vaccination.
+
+    Comparing BNT162b2mRNA and mRNA-1273 the difference in sick leave was not significant
+    after the second dose, but mRNA-1273 induced a significantly longer and more frequent
+    sick leave after the third.
+
+    Interpretation A considerable number of HCWs have been on sick leave after COVID-19
+    vaccination, staff absences increase with each additional dose, depend on the
+    vaccine, and vary between HCWs’ gender, and profession.
+
+    In the light of further COVID-19 infection waves and booster vaccinations, there
+    is a risk of additional staff shortages due to post-vaccination inability to work,
+    which could acutely overload healthcare systems and jeopardise patient care.'
+  - 'passage: title: Plasma proteomic profiling in postural orthostatic tachycardia
+    syndrome (POTS) reveals new disease pathways abstract: Abstract Postural orthostatic
+    tachycardia syndrome (POTS) is a cardiovascular autonomic disorder characterized
+    by excessive heart rate increase on standing, leading to debilitating symptoms
+    with limited therapeutic possibilities.
+
+    Proteomics is a large-scale study of proteins that enables a systematic unbiased
+    view on disease and health, allowing stratification of patients based on their
+    protein background.
+
+    The aim of the present study was to determine plasma protein biomarkers of POTS
+    and to reveal proteomic pathways differentially regulated in POTS.
+
+    We performed an age- and sex-matched, case–control study in 130 individuals (case–control
+    ratio 1:1) including POTS and healthy controls.
+
+    Mean age in POTS was 30 ± 9.
+
+    8 years (84.
+
+    6% women) versus controls 31 ± 9.
+
+    8 years (80.
+
+    0% women).
+
+    We analyzed plasma proteins using data-independent acquisition (DIA) mass spectrometry.
+
+    Pathway analysis of significantly differently expressed proteins was executed
+    using a cutoff log2 fold change set to 1.
+
+    2 and false discovery rate (p-value) of < 0.
+
+    05.
+
+    A total of 393 differential plasma proteins were identified.
+
+    Label-free quantification of DIA-data identified 30 differentially expressed proteins
+    in POTS compared with healthy controls.
+
+    Pathway analysis identified the strongest network interactions particularly for
+    proteins involved in thrombogenicity and enhanced platelet activity, but also
+    inflammation, cardiac contractility and hypertrophy, and increased adrenergic
+    activity.'
+- source_sentence: 'query: Nachträglich hat bei uns in FRA das Testen von Reisenden
+    mit AG‑Test und fachmännischem NP‑Abstrich ausgezeichnet funktioniert: SARS‑CoV‑2
+    screening strategies for returning international travellers: Evaluation of a rapid
+    antigen test approach.'
+  sentences:
+  - 'passage: title: SARS-CoV-2 screening strategies for returning international travellers:
+    Evaluation of a rapid antigen test approach abstract: International travel poses
+    the risk of importing SARS-CoV-2 infections and introducing new viral variants
+    into the country of destination.
+
+    Established measures include mandatory quarantine with the opportunity to abbreviate
+    it with a negative rapid antigen test (RAT).
+
+    A total of 1,488 returnees were tested for SARS-CoV-2 with both PCR and RAT no
+    earlier than 5 days after arrival.
+
+    We assessed the sensitivity and specificity of the RAT.
+
+    Positive samples were evaluated for infectivity in vitro in a cell culture outgrowth
+    assay.
+
+    We tracked if participants who tested negative were reported positive within 2
+    weeks of the initial test.
+
+    Potential infectiousness was determined based on symptom onset analysis, resulting
+    in a sensitivity of the antigen test of 89% in terms of infectivity.
+
+    The specificity was 100%.
+
+    All positive outgrowth assays were preceded by a positive RAT, indicating that
+    all participants with proven in vitro infectivity were correctly identified.
+
+    None of the negative participants tested positive during the follow-up.
+
+    RAT no earlier than the 5th day after arrival was a reliable method for detecting
+    infectious travellers and can be recommended as an appropriate method for managing
+    SARS-CoV-2 travel restrictions.
+
+    Compliance to the regulations and a high standard of test quality must be ensured.'
+  - 'passage: title: Persistence and Evolution of SARS-CoV-2 in an Immunocompromised
+    Host abstract: SARS-CoV-2 in an Immunocompromised Host This letter describes an
+    immunocompromised patient who had persistent infection with SARS-CoV-2 over a
+    period of months, despite several courses of remdesivi.'
+  - 'passage: title: Imperfect Vaccination Can Enhance the Transmission of Highly
+    Virulent Pathogens abstract: Could some vaccines drive the evolution of more virulent
+    pathogens?
+
+    Conventional wisdom is that natural selection will remove highly lethal pathogens
+    if host death greatly reduces transmission.
+
+    Vaccines that keep hosts alive but still allow transmission could thus allow very
+    virulent strains to circulate in a population.
+
+    Here we show experimentally that immunization of chickens against Marek''s disease
+    virus enhances the fitness of more virulent strains, making it possible for hyperpathogenic
+    strains to transmit.
+
+    Immunity elicited by direct vaccination or by maternal vaccination prolongs host
+    survival but does not prevent infection, viral replication or transmission, thus
+    extending the infectious periods of strains otherwise too lethal to persist.
+
+    Our data show that anti-disease vaccines that do not prevent transmission can
+    create conditions that promote the emergence of pathogen strains that cause more
+    severe disease in unvaccinated hosts.'
+- source_sentence: 'query: @zoeharcombe I''ll reply it. The "vaccines" have no health
+    benefit and only cause harm. Pharma company trial data says they don''t work.
+    A straightforward calculation of absolute risk from the Pfizer trial data = .04%
+    effectiveness for severe cases, which is essentially zero.'
+  sentences:
+  - 'passage: Among 10 cases of severe Covid-19 with onset after the first dose, 9
+    occurred in placebo recipients and 1 in a BNT162b2 recipient.
+
+
+    The safety profile of BNT162b2 was characterized by short-term, mild-to-moderate
+    pain at the injection site, fatigue, and headache.
+
+    The incidence of serious adverse events was low and was similar in the vaccine
+    and placebo groups.
+
+    ConclusionsA two-dose regimen of BNT162b2 conferred 95% protection against Covid-19
+    in persons 16 years of age or older.
+
+    Safety over a median of 2 months was similar to that of other viral vaccines.
+
+    (Funded by BioNTech and Pfizer; ClinicalTrials.
+
+    gov number, NCT04368728.'
+  - 'passage: title: Brain lesions disrupting addiction map to a common human brain
+    circuit abstract: Abstract Drug addiction is a public health crisis for which
+    new treatments are urgently needed.
+
+    In rare cases, regional brain damage can lead to addiction remission.
+
+    These cases may be used to identify therapeutic targets for neuromodulation.
+
+    We analyzed two cohorts of patients addicted to smoking at the time of focal brain
+    damage (cohort 1 n = 67; cohort 2 n = 62).
+
+    Lesion locations were mapped to a brain atlas and the brain network functionally
+    connected to each lesion location was computed using human connectome data ( n
+    = 1,000).
+
+    Associations with addiction remission were identified.
+
+    Generalizability was assessed using an independent cohort of patients with focal
+    brain damage and alcohol addiction risk scores ( n = 186).
+
+    Specificity was assessed through comparison to 37 other neuropsychological variables.
+
+    Lesions disrupting smoking addiction occurred in many different brain locations
+    but were characterized by a specific pattern of brain connectivity.
+
+    This pattern involved positive connectivity to the dorsal cingulate, lateral prefrontal
+    cortex, and insula and negative connectivity to the medial prefrontal and temporal
+    cortex.
+
+    This circuit was reproducible across independent lesion cohorts, associated with
+    reduced alcohol addiction risk, and specific to addiction metrics.
+
+    Hubs that best matched the connectivity profile for addiction remission were the
+    paracingulate gyrus, left frontal operculum, and medial fronto-polar cortex.
+
+    We conclude that brain lesions disrupting addiction map to a specific human brain
+    circuit and that hubs in this circuit provide testable targets for therapeutic
+    neuromodulation.'
+  - 'passage: title: Imperfect Vaccination Can Enhance the Transmission of Highly
+    Virulent Pathogens abstract: Could some vaccines drive the evolution of more virulent
+    pathogens?
+
+    Conventional wisdom is that natural selection will remove highly lethal pathogens
+    if host death greatly reduces transmission.
+
+    Vaccines that keep hosts alive but still allow transmission could thus allow very
+    virulent strains to circulate in a population.
+
+    Here we show experimentally that immunization of chickens against Marek''s disease
+    virus enhances the fitness of more virulent strains, making it possible for hyperpathogenic
+    strains to transmit.
+
+    Immunity elicited by direct vaccination or by maternal vaccination prolongs host
+    survival but does not prevent infection, viral replication or transmission, thus
+    extending the infectious periods of strains otherwise too lethal to persist.
+
+    Our data show that anti-disease vaccines that do not prevent transmission can
+    create conditions that promote the emergence of pathogen strains that cause more
+    severe disease in unvaccinated hosts.'
+pipeline_tag: sentence-similarity
+library_name: sentence-transformers
+metrics:
+- cosine_accuracy@1
+- cosine_accuracy@3
+- cosine_accuracy@5
+- cosine_accuracy@10
+- cosine_precision@1
+- cosine_precision@3
+- cosine_precision@5
+- cosine_precision@10
+- cosine_recall@1
+- cosine_recall@3
+- cosine_recall@5
+- cosine_recall@10
+- cosine_ndcg@10
+- cosine_mrr@10
+- cosine_map@100
+model-index:
+- name: SentenceTransformer based on intfloat/multilingual-e5-large
+  results:
+  - task:
+      type: information-retrieval
+      name: Information Retrieval
+    dataset:
+      name: 10 percent dev split
+      type: 10-percent-dev-split
+    metrics:
+    - type: cosine_accuracy@1
+      value: 0.52
+      name: Cosine Accuracy@1
+    - type: cosine_accuracy@3
+      value: 0.7215584415584415
+      name: Cosine Accuracy@3
+    - type: cosine_accuracy@5
+      value: 0.7937662337662338
+      name: Cosine Accuracy@5
+    - type: cosine_accuracy@10
+      value: 0.8493506493506493
+      name: Cosine Accuracy@10
+    - type: cosine_precision@1
+      value: 0.52
+      name: Cosine Precision@1
+    - type: cosine_precision@3
+      value: 0.24051948051948052
+      name: Cosine Precision@3
+    - type: cosine_precision@5
+      value: 0.15875324675324673
+      name: Cosine Precision@5
+    - type: cosine_precision@10
+      value: 0.08493506493506493
+      name: Cosine Precision@10
+    - type: cosine_recall@1
+      value: 0.52
+      name: Cosine Recall@1
+    - type: cosine_recall@3
+      value: 0.7215584415584415
+      name: Cosine Recall@3
+    - type: cosine_recall@5
+      value: 0.7937662337662338
+      name: Cosine Recall@5
+    - type: cosine_recall@10
+      value: 0.8493506493506493
+      name: Cosine Recall@10
+    - type: cosine_ndcg@10
+      value: 0.6862946585295414
+      name: Cosine Ndcg@10
+    - type: cosine_mrr@10
+      value: 0.6337252112966393
+      name: Cosine Mrr@10
+    - type: cosine_map@100
+      value: 0.6389157727076894
+      name: Cosine Map@100
+---
+
+# SentenceTransformer based on intfloat/multilingual-e5-large
+
+This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large). It maps sentences & paragraphs to a 1024-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:** [intfloat/multilingual-e5-large](https://huggingface.co/intfloat/multilingual-e5-large) <!-- at revision 3d7cfbdacd47fdda877c5cd8a79fbcc4f2a574f3 -->
+- **Maximum Sequence Length:** 256 tokens
+- **Output Dimensionality:** 1024 dimensions
+- **Similarity Function:** Cosine Similarity
+<!-- - **Training Dataset:** Unknown -->
+<!-- - **Language:** Unknown -->
+<!-- - **License:** Unknown -->
+
+### Model Sources
+
+- **Documentation:** [Sentence Transformers Documentation](https://sbert.net)
+- **Repository:** [Sentence Transformers on GitHub](https://github.com/huggingface/sentence-transformers)
+- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers)
+
+### Full Model Architecture
+
+```
+SentenceTransformer(
+  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False, 'architecture': 'PeftModelForFeatureExtraction'})
+  (1): Pooling({'word_embedding_dimension': 1024, '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})
+  (2): Normalize()
+)
+```
+
+## Usage
+
+### Direct Usage (Sentence Transformers)
+
+First install the Sentence Transformers library:
+
+```bash
+pip install -U sentence-transformers
+```
+
+Then you can load this model and run inference.
+```python
+from sentence_transformers import SentenceTransformer
+
+# Download from the 🤗 Hub
+model = SentenceTransformer("songphucn7/me5-checkthat-task1-v2")
+# Run inference
+sentences = [
+    'query: @zoeharcombe I\'ll reply it. The "vaccines" have no health benefit and only cause harm. Pharma company trial data says they don\'t work. A straightforward calculation of absolute risk from the Pfizer trial data = .04% effectiveness for severe cases, which is essentially zero.',
+    'passage: Among 10 cases of severe Covid-19 with onset after the first dose, 9 occurred in placebo recipients and 1 in a BNT162b2 recipient.\n\nThe safety profile of BNT162b2 was characterized by short-term, mild-to-moderate pain at the injection site, fatigue, and headache.\nThe incidence of serious adverse events was low and was similar in the vaccine and placebo groups.\nConclusionsA two-dose regimen of BNT162b2 conferred 95% protection against Covid-19 in persons 16 years of age or older.\nSafety over a median of 2 months was similar to that of other viral vaccines.\n(Funded by BioNTech and Pfizer; ClinicalTrials.\ngov number, NCT04368728.',
+    "passage: title: Imperfect Vaccination Can Enhance the Transmission of Highly Virulent Pathogens abstract: Could some vaccines drive the evolution of more virulent pathogens?\nConventional wisdom is that natural selection will remove highly lethal pathogens if host death greatly reduces transmission.\nVaccines that keep hosts alive but still allow transmission could thus allow very virulent strains to circulate in a population.\nHere we show experimentally that immunization of chickens against Marek's disease virus enhances the fitness of more virulent strains, making it possible for hyperpathogenic strains to transmit.\nImmunity elicited by direct vaccination or by maternal vaccination prolongs host survival but does not prevent infection, viral replication or transmission, thus extending the infectious periods of strains otherwise too lethal to persist.\nOur data show that anti-disease vaccines that do not prevent transmission can create conditions that promote the emergence of pathogen strains that cause more severe disease in unvaccinated hosts.",
+]
+embeddings = model.encode(sentences)
+print(embeddings.shape)
+# [3, 1024]
+
+# Get the similarity scores for the embeddings
+similarities = model.similarity(embeddings, embeddings)
+print(similarities)
+# tensor([[1.0000, 0.5866, 0.3283],
+#         [0.5866, 1.0000, 0.1368],
+#         [0.3283, 0.1368, 1.0000]])
+```
+
+<!--
+### Direct Usage (Transformers)
+
+<details><summary>Click to see the direct usage in Transformers</summary>
+
+</details>
+-->
+
+<!--
+### Downstream Usage (Sentence Transformers)
+
+You can finetune this model on your own dataset.
+
+<details><summary>Click to expand</summary>
+
+</details>
+-->
+
+<!--
+### Out-of-Scope Use
+
+*List how the model may foreseeably be misused and address what users ought not to do with the model.*
+-->
+
+## Evaluation
+
+### Metrics
+
+#### Information Retrieval
+
+* Dataset: `10-percent-dev-split`
+* Evaluated with [<code>InformationRetrievalEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.InformationRetrievalEvaluator)
+
+| Metric              | Value      |
+|:--------------------|:-----------|
+| cosine_accuracy@1   | 0.52       |
+| cosine_accuracy@3   | 0.7216     |
+| cosine_accuracy@5   | 0.7938     |
+| cosine_accuracy@10  | 0.8494     |
+| cosine_precision@1  | 0.52       |
+| cosine_precision@3  | 0.2405     |
+| cosine_precision@5  | 0.1588     |
+| cosine_precision@10 | 0.0849     |
+| cosine_recall@1     | 0.52       |
+| cosine_recall@3     | 0.7216     |
+| cosine_recall@5     | 0.7938     |
+| cosine_recall@10    | 0.8494     |
+| **cosine_ndcg@10**  | **0.6863** |
+| cosine_mrr@10       | 0.6337     |
+| cosine_map@100      | 0.6389     |
+
+<!--
+## Bias, Risks and Limitations
+
+*What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.*
+-->
+
+<!--
+### Recommendations
+
+*What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.*
+-->
+
+## Training Details
+
+### Training Dataset
+
+#### Unnamed Dataset
+
+* Size: 17,319 training samples
+* Columns: <code>sentence_0</code> and <code>sentence_1</code>
+* Approximate statistics based on the first 1000 samples:
+  |         | sentence_0                                                                         | sentence_1                                                                          |
+  |:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------|
+  | type    | string                                                                             | string                                                                              |
+  | details | <ul><li>min: 21 tokens</li><li>mean: 59.2 tokens</li><li>max: 136 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 205.0 tokens</li><li>max: 256 tokens</li></ul> |
+* Samples:
+  | sentence_0                                                                                                                                                                                                                                                                                         | sentence_1                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      |
+  |:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+  | <code>query: @user Baloney. Natural immunity is hands down better, and vaccinated people are ending up in the hospital.</code>                                                                                                                                                                     | <code>passage: title: Longitudinal analysis shows durable and broad immune memory after SARS-CoV-2 infection with persisting antibody responses and memory B and T cells abstract: Ending the COVID-19 pandemic will require long-lived immunity to SARS-CoV-2.<br>Here, we evaluate 254 COVID-19 patients longitudinally up to 8 months and find durable broad-based immune responses.<br>SARS-CoV-2 spike binding and neutralizing antibodies exhibit a bi-phasic decay with an extended half-life of >200 days suggesting the generation of longer-lived plasma cells.<br>SARS-CoV-2 infection also boosts antibody titers to SARS-CoV-1 and common betacoronaviruses.<br>In addition, spike-specific IgG+ memory B cells persist, which bodes well for a rapid antibody response upon virus re-exposure or vaccination.<br>Virus-specific CD4+ and CD8+ T cells are polyfunctional and maintained with an estimated half-life of 200 days.<br>Interestingly, CD4+ T cell responses equally target several SARS-CoV-2 proteins, whereas the CD8+ T cell respo...</code>                      |
+  | <code>query: @Alexand64744343 Meta examen tests #lyme + élevé niveau de preuve scientifique ! : rendement « 53.9% for synthetic C6 peptide ELISA tests & 53.7% when the two-tier methodology was used » Une véritable loterie, 1 cas sur 2 détecté mais persistez à vociférer par ignorance</code> | <code>passage: title: Commercial test kits for detection of Lyme borreliosis: a meta-analysis of test accuracy abstract: The clinical diagnosis of Lyme borreliosis can be supported by various test methodologies; test kits are available from many manufacturers.<br>Literature searches were carried out to identify studies that reported characteristics of the test kits.<br>Of 50 searched studies, 18 were included where the tests were commercially available and samples were proven to be positive using serology testing, evidence of an erythema migrans rash, and/or culture.<br>Additional requirements were a test specificity of ≥85% and publication in the last 20 years.<br>The weighted mean sensitivity for all tests and for all samples was 59.<br>5%.<br>Individual study means varied from 30.<br>6% to 86.<br>2%.<br>Sensitivity for each test technology varied from 62.<br>4% for Western blot kits, and 62.<br>3% for enzyme-linked immunosorbent assay tests, to 53.<br>9% for synthetic C6 peptide ELISA tests and 53.<br>7% when the two-tier meth...</code> |
+  | <code>query: 28 Les systèmes de séquençage haut débit qui servent à la production des banques comme celles du papier de Jaenisch produisent des chimères artefactuelles lors de la PCR. C’est bien connu. Discuté dans :</code>                                                                    | <code>passage: title: A Survey of Virus Recombination Uncovers Canonical Features of Artificial Chimeras Generated During Deep Sequencing Library Preparation abstract: Abstract Chimeric reads can be generated by in vitro recombination during the preparation of high-throughput sequencing libraries.<br>Our attempt to detect biological recombination between the genomes of dengue virus (DENV; +ssRNA genome) and its mosquito host using the Illumina Nextera sequencing library preparation kit revealed that most, if not all, detected host–virus chimeras were artificial.<br>Indeed, these chimeras were not more frequent than with control RNA from another species (a pillbug), which was never in contact with DENV RNA prior to the library preparation.<br>The proportion of chimera types merely reflected those of the three species among sequencing reads.<br>Chimeras were frequently characterized by the presence of 1-20 bp microhomology between recombining fragments.<br>Within-species chimeras mostly involved fragments in...</code>                         |
+* Loss: [<code>MultipleNegativesRankingLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#multiplenegativesrankingloss) with these parameters:
+  ```json
+  {
+      "scale": 20.0,
+      "similarity_fct": "cos_sim",
+      "gather_across_devices": false,
+      "directions": [
+          "query_to_doc"
+      ],
+      "partition_mode": "joint",
+      "hardness_mode": null,
+      "hardness_strength": 0.0
+  }
+  ```
+
+### Training Hyperparameters
+#### Non-Default Hyperparameters
+
+- `per_device_train_batch_size`: 32
+- `num_train_epochs`: 10
+- `eval_strategy`: steps
+- `per_device_eval_batch_size`: 32
+- `multi_dataset_batch_sampler`: round_robin
+
+#### All Hyperparameters
+<details><summary>Click to expand</summary>
+
+- `per_device_train_batch_size`: 32
+- `num_train_epochs`: 10
+- `max_steps`: -1
+- `learning_rate`: 5e-05
+- `lr_scheduler_type`: linear
+- `lr_scheduler_kwargs`: None
+- `warmup_steps`: 0
+- `optim`: adamw_torch_fused
+- `optim_args`: None
+- `weight_decay`: 0.0
+- `adam_beta1`: 0.9
+- `adam_beta2`: 0.999
+- `adam_epsilon`: 1e-08
+- `optim_target_modules`: None
+- `gradient_accumulation_steps`: 1
+- `average_tokens_across_devices`: True
+- `max_grad_norm`: 1
+- `label_smoothing_factor`: 0.0
+- `bf16`: False
+- `fp16`: False
+- `bf16_full_eval`: False
+- `fp16_full_eval`: False
+- `tf32`: None
+- `gradient_checkpointing`: False
+- `gradient_checkpointing_kwargs`: None
+- `torch_compile`: False
+- `torch_compile_backend`: None
+- `torch_compile_mode`: None
+- `use_liger_kernel`: False
+- `liger_kernel_config`: None
+- `use_cache`: False
+- `neftune_noise_alpha`: None
+- `torch_empty_cache_steps`: None
+- `auto_find_batch_size`: False
+- `log_on_each_node`: True
+- `logging_nan_inf_filter`: True
+- `include_num_input_tokens_seen`: no
+- `log_level`: passive
+- `log_level_replica`: warning
+- `disable_tqdm`: False
+- `project`: huggingface
+- `trackio_space_id`: trackio
+- `eval_strategy`: steps
+- `per_device_eval_batch_size`: 32
+- `prediction_loss_only`: True
+- `eval_on_start`: False
+- `eval_do_concat_batches`: True
+- `eval_use_gather_object`: False
+- `eval_accumulation_steps`: None
+- `include_for_metrics`: []
+- `batch_eval_metrics`: False
+- `save_only_model`: False
+- `save_on_each_node`: False
+- `enable_jit_checkpoint`: False
+- `push_to_hub`: False
+- `hub_private_repo`: None
+- `hub_model_id`: None
+- `hub_strategy`: every_save
+- `hub_always_push`: False
+- `hub_revision`: None
+- `load_best_model_at_end`: False
+- `ignore_data_skip`: False
+- `restore_callback_states_from_checkpoint`: False
+- `full_determinism`: False
+- `seed`: 42
+- `data_seed`: None
+- `use_cpu`: False
+- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
+- `parallelism_config`: None
+- `dataloader_drop_last`: False
+- `dataloader_num_workers`: 0
+- `dataloader_pin_memory`: True
+- `dataloader_persistent_workers`: False
+- `dataloader_prefetch_factor`: None
+- `remove_unused_columns`: True
+- `label_names`: None
+- `train_sampling_strategy`: random
+- `length_column_name`: length
+- `ddp_find_unused_parameters`: None
+- `ddp_bucket_cap_mb`: None
+- `ddp_broadcast_buffers`: False
+- `ddp_backend`: None
+- `ddp_timeout`: 1800
+- `fsdp`: []
+- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
+- `deepspeed`: None
+- `debug`: []
+- `skip_memory_metrics`: True
+- `do_predict`: False
+- `resume_from_checkpoint`: None
+- `warmup_ratio`: None
+- `local_rank`: -1
+- `prompts`: None
+- `batch_sampler`: batch_sampler
+- `multi_dataset_batch_sampler`: round_robin
+- `router_mapping`: {}
+- `learning_rate_mapping`: {}
+
+</details>
+
+### Training Logs
+| Epoch  | Step | Training Loss | 10-percent-dev-split_cosine_ndcg@10 |
+|:------:|:----:|:-------------:|:-----------------------------------:|
+| 0.1845 | 100  | -             | 0.6380                              |
+| 0.3690 | 200  | -             | 0.6400                              |
+| 0.5535 | 300  | -             | 0.6512                              |
+| 0.7380 | 400  | -             | 0.6642                              |
+| 0.9225 | 500  | 0.8957        | 0.6640                              |
+| 1.0    | 542  | -             | 0.6626                              |
+| 1.1070 | 600  | -             | 0.6658                              |
+| 1.2915 | 700  | -             | 0.6676                              |
+| 1.4760 | 800  | -             | 0.6695                              |
+| 1.6605 | 900  | -             | 0.6719                              |
+| 1.8450 | 1000 | 0.3899        | 0.6776                              |
+| 2.0    | 1084 | -             | 0.6744                              |
+| 2.0295 | 1100 | -             | 0.6761                              |
+| 2.2140 | 1200 | -             | 0.6759                              |
+| 2.3985 | 1300 | -             | 0.6761                              |
+| 2.5830 | 1400 | -             | 0.6830                              |
+| 2.7675 | 1500 | 0.3484        | 0.6779                              |
+| 2.9520 | 1600 | -             | 0.6793                              |
+| 3.0    | 1626 | -             | 0.6762                              |
+| 3.1365 | 1700 | -             | 0.6823                              |
+| 3.3210 | 1800 | -             | 0.6831                              |
+| 3.5055 | 1900 | -             | 0.6788                              |
+| 3.6900 | 2000 | 0.3083        | 0.6821                              |
+| 3.8745 | 2100 | -             | 0.6775                              |
+| 4.0    | 2168 | -             | 0.6788                              |
+| 4.0590 | 2200 | -             | 0.6786                              |
+| 4.2435 | 2300 | -             | 0.6792                              |
+| 4.4280 | 2400 | -             | 0.6827                              |
+| 4.6125 | 2500 | 0.3033        | 0.6804                              |
+| 4.7970 | 2600 | -             | 0.6822                              |
+| 4.9815 | 2700 | -             | 0.6914                              |
+| 5.0    | 2710 | -             | 0.6880                              |
+| 5.1661 | 2800 | -             | 0.6809                              |
+| 5.3506 | 2900 | -             | 0.6853                              |
+| 5.5351 | 3000 | 0.2840        | 0.6852                              |
+| 5.7196 | 3100 | -             | 0.6844                              |
+| 5.9041 | 3200 | -             | 0.6886                              |
+| 6.0    | 3252 | -             | 0.6859                              |
+| 6.0886 | 3300 | -             | 0.6859                              |
+| 6.2731 | 3400 | -             | 0.6811                              |
+| 6.4576 | 3500 | 0.2669        | 0.6896                              |
+| 6.6421 | 3600 | -             | 0.6864                              |
+| 6.8266 | 3700 | -             | 0.6859                              |
+| 7.0    | 3794 | -             | 0.6893                              |
+| 7.0111 | 3800 | -             | 0.6907                              |
+| 7.1956 | 3900 | -             | 0.6865                              |
+| 7.3801 | 4000 | 0.2546        | 0.6831                              |
+| 7.5646 | 4100 | -             | 0.6872                              |
+| 7.7491 | 4200 | -             | 0.6893                              |
+| 7.9336 | 4300 | -             | 0.6864                              |
+| 8.0    | 4336 | -             | 0.6900                              |
+| 8.1181 | 4400 | -             | 0.6885                              |
+| 8.3026 | 4500 | 0.2518        | 0.6857                              |
+| 8.4871 | 4600 | -             | 0.6874                              |
+| 8.6716 | 4700 | -             | 0.6834                              |
+| 8.8561 | 4800 | -             | 0.6859                              |
+| 9.0    | 4878 | -             | 0.6858                              |
+| 9.0406 | 4900 | -             | 0.6844                              |
+| 9.2251 | 5000 | 0.2392        | 0.6861                              |
+| 9.4096 | 5100 | -             | 0.6874                              |
+| 9.5941 | 5200 | -             | 0.6872                              |
+| 9.7786 | 5300 | -             | 0.6858                              |
+| 9.9631 | 5400 | -             | 0.6861                              |
+| 10.0   | 5420 | -             | 0.6863                              |
+
+
+### Framework Versions
+- Python: 3.12.6
+- Sentence Transformers: 5.3.0
+- Transformers: 5.5.1
+- PyTorch: 2.11.0+cu130
+- Accelerate: 1.13.0
+- Datasets: 4.8.4
+- Tokenizers: 0.22.2
+
+## Citation
+
+### BibTeX
+
+#### Sentence Transformers
+```bibtex
+@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
+```bibtex
+@misc{oord2019representationlearningcontrastivepredictive,
+      title={Representation Learning with Contrastive Predictive Coding},
+      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
+      year={2019},
+      eprint={1807.03748},
+      archivePrefix={arXiv},
+      primaryClass={cs.LG},
+      url={https://arxiv.org/abs/1807.03748},
+}
+```
+
+<!--
+## Glossary
+
+*Clearly define terms in order to be accessible across audiences.*
+-->
+
+<!--
+## Model Card Authors
+
+*Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.*
+-->
+
+<!--
+## Model Card Contact
+
+*Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.*
+-->

adapter_config.json

@@ -0,0 +1,41 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "intfloat/multilingual-e5-large",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.18.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "query",
+    "value"
+  ],
+  "target_parameters": null,
+  "task_type": "FEATURE_EXTRACTION",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1e36ad0e4b15be93da9fe836c6e0bbe9ef25351e4a65774d7a68db082435a14b
+size 6304672

config_sentence_transformers.json

@@ -0,0 +1,14 @@
+{
+  "model_type": "SentenceTransformer",
+  "__version__": {
+    "sentence_transformers": "5.3.0",
+    "transformers": "5.5.1",
+    "pytorch": "2.11.0+cu130"
+  },
+  "prompts": {
+    "query": "",
+    "document": ""
+  },
+  "default_prompt_name": null,
+  "similarity_fn_name": "cosine"
+}

modules.json

@@ -0,0 +1,20 @@
+[
+  {
+    "idx": 0,
+    "name": "0",
+    "path": "",
+    "type": "sentence_transformers.models.Transformer"
+  },
+  {
+    "idx": 1,
+    "name": "1",
+    "path": "1_Pooling",
+    "type": "sentence_transformers.models.Pooling"
+  },
+  {
+    "idx": 2,
+    "name": "2",
+    "path": "2_Normalize",
+    "type": "sentence_transformers.models.Normalize"
+  }
+]

sentence_bert_config.json

@@ -0,0 +1,4 @@
+{
+    "max_seq_length": 256,
+    "do_lower_case": false
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a514807cffabd8abaf028cfaffe7ff0c4f60b97ea2db80c41f14172ae6b018ca
+size 17082987

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "add_prefix_space": true,
+  "backend": "tokenizers",
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "is_local": false,
+  "mask_token": "<mask>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "unk_token": "<unk>"
+}