README.md

---
base_model: sentence-transformers/all-MiniLM-L6-v2
datasets: []
language: []
library_name: sentence-transformers
metrics:
- pearson_cosine
- spearman_cosine
- pearson_manhattan
- spearman_manhattan
- pearson_euclidean
- spearman_euclidean
- pearson_dot
- spearman_dot
- pearson_max
- spearman_max
pipeline_tag: sentence-similarity
tags:
- sentence-transformers
- sentence-similarity
- feature-extraction
- generated_from_trainer
- dataset_size:28450
- loss:MultipleNegativesRankingLoss
widget:
- source_sentence: What are the five criteria that community projects must meet to
    be considered for funding by the Community Ownership Fund?
  sentences:
  - 'We want to fund community projects that do at least 1 of these 5 things:

    increase feelings of pride in, and improve perceptions of, the local area as a
    place to live

    improve social trust, cohesion, and sense of belonging

    increase local participation in community life, arts, culture, or sport

    improve local economic outcomes – including creating jobs, volunteering opportunities,
    and improving employability and skills levels in the local community

    improve social and wellbeing outcomes – including having a positive impact on
    physical and mental health of local people, and reducing loneliness and social
    isolation

    Strengthening community ownership across the UK

    The Fund will be delivered directly by the UK government to communities in England,
    Scotland, Wales, and Northern Ireland. The UK government is committed to fair
    opportunities to access funding through the Community Ownership Fund across the
    UK.

    A minimum target of spending in line with per-capita allocations has therefore
    been set in Scotland, Wales, and Northern Ireland. The Community Ownership Fund
    will target a minimum of £12.3 million in Scotland, £7.1 million in Wales, and
    £4.3 million in Northern Ireland of the total Fund over the 4 years until March
    2025.

    The design of the Fund recognises the different landscapes for community ownership
    across the UK, with different legislation in England and Wales, Scotland, and
    Northern Ireland. We have engaged widely with local stakeholders to ensure the
    Fund is effective, accessible and achieves its objectives.

    Applications will be assessed against a consistent framework. Eligibility for
    the Fund and the bidding assessment criteria are consistent in all 4 nations.

    Glossary

    Community asset

    For this fund, an asset is physical building or space. It must be used by the
    community and accessible to as many people as possible.

    Community Asset Transfer

    Across the United Kingdom, Community Asset Transfer (CAT) policy frameworks support
    the transfer of community assets from public authorities to community organisations.
    The legislation and policy contexts work slightly differently in parts of the
    United Kingdom.

    England

    Asset of community value

    In England the Localism Act 2011 introduced a right for community groups to nominate
    buildings or land to their local authority as an asset of community value.

    If the local authority agreed that the nomination met the test of being land of
    community value, the council would place the asset on a list of assets of community
    value for a period of 5 years.

    What this did was introduce a community right to bid. If the owner of a listed
    asset decided that they wish to sell the asset during the 5-year period of listing,
    then they must notify the local authority who would inform the nominating community
    group.'
  - "In designated catchments, water companies have a duty to ensure wastewater treatment\
    \ works serving a population equivalent over 2,000 meet specified nutrient removal\
    \ standards by 1 April 2030 where the designation takes effect from 25 January\
    \ 2024. For designations that take effect subsequent to that date, the upgrade\
    \ date is specified in the notice. Competent authorities (including local planning\
    \ authorities) considering planning proposals for development draining via a sewer\
    \ to a wastewater treatment works subject to the upgrade duty are required to\
    \ consider that the nutrient pollution standard will be met by the upgrade date\
    \ for the purposes of Habitats Regulations Assessments.  \nWhilst the upgrade\
    \ date under the Water Industry Act 1991 for this catchment is 16 May 2031, the\
    \ sewerage undertaker has committed to the delivery of the wastewater treatment\
    \ work upgrades by 1 April 2030. The Environment Agency has also committed to\
    \ varying Environmental Permits for the relevant wastewater treatment works so\
    \ that the permits will require compliance with the nutrient pollution standard\
    \ by 1 April 2030. ↩"
  - 'https://gcscc.ox.ac.uk/cmm-reviews#/ ↩

    World Bank, ‘Green Digital Transformation: How to Sustainably Close the Digital
    Divide and Harness Digital Tools for Climate Action’ https://openknowledge.worldbank.org/entities/
    publication/6be73f14-f899-4a6d-a26e-56d98393acf3 ↩

    Ritchie, 2020 https://ourworldindata.org/ghg-emissions-by-sector ↩

    WHO, e-waste factsheet, 2023: https://www.who.int/news-room/fact-sheets/detail/
    electronic-waste-(e-waste) ↩

    International development in a contested world: ending extreme poverty and tackling
    climate change https://www.gov.uk/government/publications/international-development-in-a-contested-world-ending-extreme-poverty-and-tackling-climate-change
    ↩

    https://www.gov.uk/government/publications/greening-government-ict-and-digitalservices-strategy-2020-2025
    ↩

    UK Government’s Department for Environment, Food & Rural Affairs ↩

    https://digitalprinciples.org/ ↩

    https://www.dynamicspectrumalliance.org/ ↩

    https://www.itu.int/itu-d/sites/partner2connect/ ↩

    https://www.govstack.global/ ↩'
- source_sentence: What specific actions is the UK government implementing as part
    of the third National Adaptation Programme (NAP3) to address the impacts of climate
    change?
  sentences:
  - '(The Thames Barrier in London, shown at low tide. Photo by mikeinlondon via Getty
    Images.)

    The government is taking action to adapt the UK to climate change. This can help
    reduce the costs from climate change impacts and make our economy and society
    more resilient.

    This page explains more about:

    climate change and adaptation

    the risks and opportunities of climate change

    what the government is doing to make sure that the UK is prepared for climate
    change – including the third National Adaptation Programme (NAP3)

    Climate change

    Our climate is changing. The main cause is human activity: in particular, burning
    fossil fuels for energy, which emits greenhouse gases into the atmosphere and
    causes the world’s temperature to rise.

    In the UK we can see the effects of climate change already. In 2022 the UK recorded
    the warmest year on record with temperatures reaching over 40°C, which had impacts
    on public health and the environment. These temperatures would not have been possible
    without climate change caused by human activity. The frequency of hotter summers
    will increase in the future, and we can expect the winters to become wetter, which
    will make flooding more likely across the UK.

    The government is taking action to limit climate change through its commitment
    to reach net zero greenhouse gas emissions by 2050. One of these actions is reducing
    our reliance on fossil fuels. Achieving ‘net zero’ in the UK and across the world
    will help to limit temperature rises in the future and reduce the level of climate
    change we need to adapt to.

    Climate adaptation

    Climate adaptation relates to actions that protect us against the impacts of climate
    change. This includes reacting to the changes we have seen already, as well as
    preparing for what will happen in the future.

    The UK government is taking steps to address the impacts of climate change to
    protect communities, our economy and the environment.

    Examples of the government’s approach to climate adaptation include:

    building new flood defences to protect against rising sea levels

    planning for more green spaces in urban areas to help keep them cool and planting
    more drought-resistant crops

    building infrastructure that can withstand expected climate impacts such as extreme
    heat and flooding

    Many of the actions in NAP3 can help to improve our standard of living too, by
    upgrading our buildings and infrastructure, improving the sustainability and productivity
    of important sectors such as agriculture and forestry, and restoring our natural
    environment.

    Climate risks and opportunities

    Climate change can lead to both risks and opportunities, although there are more
    risks than opportunities. Without measures to adapt to climate change, we would
    experience additional issues including:

    health risks

    damage to houses and infrastructure'
  - 'We will help shape an international order in which all citizens are well informed,
    able to participate in democratic processes and enjoy their rights in offline
    and online public spaces, as well as freedom of expression; and we will promote
    an information ecosystem that supports accountability and inclusive deliberative
    democracy.

    The UK commits to an open, free, global, interoperable, reliable and secure Internet;
    and to ensuring emerging tech supports, rather than erodes, the enjoyment of democracy,
    human rights and fundamental freedoms. Working collectively with international
    partners, civil society and the tech sector is critical in ensuring that the online
    world and technologies promote freedom, democracy and inclusion, and protect human
    rights and fundamental freedoms.

    We will strengthen our collaboration in the multi-stakeholder spaces that support
    digital democracy. We will enhance our advisory support to the Freedom Online
    Coalition (FOC) and will bid to continue as a member of the FOC Steering Committee
    and to maintain our role as co-chairs of the Taskforce on Internet Shutdowns (TFIS).

    We will support our overseas network to better understand the threat posed by
    information disorder through digital platforms. In doing so, we will identify
    international best practice and increase our understanding of information disorder
    in elections, independent media as well as gendered disinformation impacts on
    women’s political empowerment and participation in electoral processes.

    We will champion the importance of a vibrant, independent, and pluralistic civic
    space online and offline, where people can exercise their freedoms. We will work
    in collaboration with other donors, civil society, academia and the private sector
    to leverage the opportunities and mitigate the risks that digital transformation
    provides for civil society and civic space.

    We will support open and accountable use of emerging digital technologies, especially
    the need for democratic and human rights safeguards. This includes grant support
    for the Open Government Partnership to help enable open and accountable use of
    emerging digital technologies by driving digital governance reforms in 10 countries
    (Ghana, Indonesia, Kenya, Nigeria, Dominic Republic, Armenia, Colombia, Zambia,
    the Philippines and Ukraine), accelerating collective action and norm-raising
    on digital governance and increasing impact through better connection between
    global pledges and country action.

    Chapter 3 – Digital inclusion: leaving no one behind in a digital world

    The benefits of digital transformation are not evenly distributed. A third of
    the world’s population is offline, and that is concentrated within the poorest
    and most marginalised groups.'
  - 'Estimated one-off impact on administrative burden (£ million)

    One-off impact (£ million) £30,000 to £50,000 threshold Above £50,000 threshold
    Total mandated population above £30,000

    Costs 338 223 561

    Savings — — —

    Estimated continuing impact on administrative burden (£ million)

    Continuing average annual impact (£ million) £30,000 to £50,000 threshold Above
    £50,000 threshold Total mandated population above £30,000

    Costs 110 90 201

    Savings 2 3 5

    Net impact on annual administrative burden +108 +88 +196

    Numbers do not sum due to rounding.

    Operational impact (£ million) (HMRC or other)

    There will be both IT and resource costs for HMRC in developing, applying, and
    policing this measure, and in updating guidance.

    HMRC IT and non-IT costs for this next phase of MTD expansion are expected to
    be in the region of £0.5bn to the end of March 2028.

    Other impacts

    HMRC is required to consider the justice impact test and rural proofing measures
    in relation to their impacts on rural communities and the justice system.

    HMRC’s assessments suggest any impact is likely to be negligible. Mitigations
    are in place for those whose rural location impacts their internet access to the
    point where it is not feasible to operate MTD, as discussed in the ‘Equalities
    impacts’ section.

    This measure does not fall within the scope of the environmental principles duty.

    Other impacts have been considered and none have been identified.

    Monitoring and evaluation

    HMRC’s communications programme includes work to build software developer, agent
    and taxpayer readiness, to promote inclusion in the large-scale public beta testing
    programme beginning in 2025 and encourage voluntary early adoption of MTD for
    ITSA.

    HMRC is committed to monitoring and formally evaluating the impact of MTD for
    ITSA, including both customer and revenue impacts. This will build on HMRC’s track
    record in successfully evaluating MTD for VAT and publishing the findings. Independent
    social research will be undertaken both before and after MTD for ITSA is introduced
    to gather evidence of customer impacts and behaviour change. Self Assessment data
    will be used to monitor take-up and estimate additional tax revenue due to MTD.
    The evaluation will take until at least 2029, when all data for the 2027 to 28
    tax year becomes available for analysis.

    Further advice'
- source_sentence: Who are the joint leaders of the new Anti-social Behaviour Taskforce
    responsible for overseeing the implementation and delivery of the action plan?
  sentences:
  - '80. It is also vital that we measure the overall success of this plan in tackling
    anti-social behaviour to ensure that it is meeting the commitments we have set
    out. We will assess the impact of our proposals on both communities’ experience
    and perceptions of anti-social behaviour and their effectiveness in tackling it.
    To achieve this, we will draw from the wide range of data enhancements outlined
    throughout this plan, alongside wider measures, to monitor and evaluate its success
    and to further inform our understanding of what works in driving down anti-social
    behaviour.

    81. We will oversee the implementation and delivery to this action plan with a
    new Anti-social Behaviour Taskforce jointly led by the Home Secretary and the
    Secretary of State for Levelling Up that will bring together national and local
    partners, with a sole focus of addressing anti-social behaviour and restoring
    pride in place in communities.

    Home Office. Anti-social behaviour: impacts on individuals and local communities.
    2023 ↩

    Home Office. Guidance: Anti-social behaviour principles. 2022. ↩

    Home Office. Anti-social behaviour: impacts on individuals and local communities.
    2023. ↩

    YouGov. Anti-Social Behaviour. 2023. ↩

    A legal definition of ASB can be found in the Anti-Social Behaviour Act 2014:
    a) conduct that has caused, or is likely to cause, harassment, alarm or distress
    to any person, b) conduct capable of causing nuisance or annoyance to a person
    in relation to that person’s occupation of residential premises, or c) conduct
    capable of causing housing-related nuisance or annoyance to any person. ↩

    Ipsos. Ipsos Levelling Up Index: Levelling up Panel. 2022. ↩

    Public First. Levelling Up Poll. 2021. ↩

    Office for National Statistics. Crime in England and Wales: Other related tables
    . 2022. ↩

    Office for National Statistics. Crime Survey for England and Wales (CSEW) estimates
    of personal and household crime, anti-social behaviour, and public perceptions,
    by police force area, year ending September 2022. ↩

    Office for National Statistics. Crime in England and Wales: Police Force Area
    data tables. 2023. Office for National Statistics. Crime in England and Wales:
    Other related tables. 2023. Office for National Statistics. Crime in England and
    Wales: Annual Trend and Demographic Tables. 2022. ↩'
  - '323. Similarly, DCMS Ministers in both Houses of Parliament expressed at the
    dispatch box their disappointment about the proposed changes to BBC local radio
    services. There have also been several instances over the Charter period where
    a lack of effective transparency in engaging the public has been highlighted in
    the media and by Parliamentarians. For example, the BBC’s failure to explain how
    it was dealing with complaints about the anti-semitic incident on a bus on Oxford
    Street at the end of 2021 in the face of significant public pressure received
    widespread media coverage. The announcement of the closure of BBC Singers led
    to Parliamentary discussions and media reports raising concerns about how the
    decision had been made and communicated, including internally within the BBC.

    The government’s response

    324. When considering how the BBC communicates with audiences, it is our view
    that the BBC should be held to a higher standard than other organisations given
    the extent of its public funding. This higher standard needs to go beyond publication
    of more data and information, to straightforward and open communication with audiences.
    The BBC Board has overall responsibility for ensuring that the BBC communicates
    changes that have an impact on audiences effectively with those audiences. This
    has to be accompanied by equally effective communication with its workforce. Evidence
    received indicates that the BBC has not always achieved this.

    7.1 We recommend that the BBC continues to learn from recent experiences where
    announcements about service changes have led to criticism about the BBC’s approach
    to transparency.

    7.2 We also recommend that the BBC publishes details of its strategy for communicating
    with audiences which explains improvements to its communications approach already
    made, but also how it identifies any changes needed so that audiences and staff
    can be confident that future service changes and their impact will be explained
    clearly.

    Understanding audience needs

    What we learnt

    325. During evidence gathering, many stakeholders made proposals regarding how
    the BBC could improve its transparency in specific ways to help audiences hold
    it to account. All of these proposals related to individual specific themes in
    previous chapters. Ofcom’s research suggests that there are perception issues
    with the BBC’s impartiality that more effective transparency could help address.

    The government’s response

    326. It is important that licence fee payers do not just have the opportunity
    to shape the services that the BBC provides, but that they also have the opportunity
    to tell the BBC how they would like the BBC to be more transparent.'
  - '67. Building on our Fraud Plan, DWP is investing £70 million between 2022/23
    and 2024/25 in advanced analytics to tackle fraud and error, which it expects
    will help it to generate savings of around £1.6 billion by 2030/31[footnote 24].

    68. Investing in advanced analytics, such as machine learning, is essential to
    enable the public sector to keep up with offenders. Sophisticated crimminals already
    utilise such tools to analyse large amounts of data to exploit existing weaknesses
    and vulnerabilities in public sector systems. In DWP these tools can play a crucial
    role in detecting and preventing fraudulent activities in DWPs benefit systems.
    Going forward we want to maximise the benefits that advanced analytics and machine
    learning can offer.

    69. Where these tools are used to assist in the prevention and detection of fraud,
    DWP always ensures appropriate safeguards are in place to ensure the proportionate,
    ethical, and lawful use of data with human input. In decision making, any final
    decision will always be made by a member of DWP staff and DWP seeks to ensure
    compliance using internal monitoring protocols. DWPs Personal Information Charter
    sets out in more detail how the Department uses these tools, as well as Artificial
    Intelligence and automated decision making.

    Continuous improvement to Universal Credit (UC)

    70. As we complete the Move to UC, the Department’s spending on UC alone is forecast
    to double (relative to 2022/23 in nominal terms) to reach over £85 billion by
    2028/29[footnote 25].

    71. We are constantly improving UC to reduce fraud and error and to ensure the
    right support reaches the right people.

    72. Building on our previous Fraud Plan our UC Continuous Improvement plan brings
    together multi-disciplinary teams to look at the largest areas of loss within
    UC and considers how we can improve our processes to reduce these.

    73. These teams focus on understanding the root-causes and scale of the losses,
    design and test solutions with a view to implementing them more widely if the
    tests are successful. The implementation of these solutions may involve changes
    to policy, improvements to the operation of UC service or greater use of data
    and automation to prevent the fraud.'
- source_sentence: What is the date and time of the next meeting?
  sentences:
  - 'Defra is working with the British Standards Institution (BSI) to develop a suite
    of nature investment standards that will support best practice standardisation
    of methodologies with regards to best practices for assessing the baseline, monitoring,
    and verifying the delivery of nature-based carbon removals. This will be critical
    for the purposes of supplying and selling credits into nature markets, and for
    quantifying within value chain mitigation of environmental impacts. These standards
    will build on and aim to align with the work of international integrity initiatives,
    including the Integrity Council for Voluntary Carbon Markets (ICVCM) and the Voluntary
    Carbon Markets Initiative (VCMI).

    As part of this programme, BSI is developing the ‘Nature markets - Overarching
    principles and framework’, which will apply to nature-based environmental improvement
    projects and the quantification of ecosystem services. These principles will set
    the basis by which nature markets can be more effectively designed and governed.
    A first draft of the BSI Flex 701 standard was published for consultation in March
    2024.

    Further to this, BSI will be developing more specific thematic and market specific
    standards to follow over the course of 2024 to 2025, for example, for nature-based
    carbon and biodiversity. This will include a certification mechanism to allow
    methodologies which meet these standards to become certified as offering high
    integrity.

    1.2 A standardised approach to product level impact quantification

    Increasingly, businesses are seeing the benefits of communicating product level
    impact data to consumers and other businesses in the supply chain. Product level
    accounting can help improve understanding of the impacts of specific products
    and supply chains to inform changes at the supplier and product level to reduce
    impacts. Product level data can also enable more accurate reporting of company
    impacts from the ‘bottom-up’, by summing up the impact of all products sold by
    the company, in addition to any energy use or emissions on site.

    Product level impact data is generated through lifecycle assessments (LCAs). Although
    there are many commonalities between Scope 3 and product carbon footprinting,
    there are a number of practical and methodological differences summarised in section
    4.1 of the WRAP Protocol.

    Relevant priorities

    1.3 – A standardised product level accounting method (including multi-metric approach)

    Developing a product level accounting method'
  - 'To enable efficient and extensive use of genomic AMR data, the design and implementation
    of data handling solutions will be explored. The design should accommodate complexities
    such as AMR outbreaks caused by the same AMR-causing mobile genetic element transferred
    among different pathogen species, or longer-term trends in AMR epidemiology. These
    should provide new or use existing open standards, for the handling of AMR-related
    information, to facilitate working with international partners and allow convenient
    and effective querying for surveillance and response planning. Few countries offer
    large scale sequencing and analysis of AMR associated isolates so UK data would
    provide vital insight into the molecular epidemiology of these infections and
    position the UK to exploit the knowledge these new methods can provide.

    Theme 2 - Optimising the use of antimicrobials

    Outcome 4 - Antimicrobial stewardship and disposal

    By 2029, the UK has strengthened antimicrobial stewardship and diagnostic stewardship
    by improved targeting of antimicrobials and diagnostic tools for humans, animals
    and plants, and improved the disposal of antimicrobials, informed by the right
    data, risk stratification and guidance.

    This outcome has:

    3 commitments:

    clinical decision support

    appropriate prescribing and disposal

    behavioural interventions

    2 human health targets (see appendix B):

    target 4a: by 2029, we aim to reduce total antibiotic use in human populations
    by 5% from the 2019 baseline

    target 4b: by 2029, we aim to achieve 70% of total use of antibiotics from the
    Access category (new UK category) across the human healthcare system

    While all use of antimicrobials drives AMR, there is an opportunity to reduce
    inappropriate use of antimicrobials occurring, for example, when antimicrobials
    are taken when they are not needed, or when taken for longer than necessary.

    According to the National Institute for Health and Care Excellence’s NICE guideline
    (NG15):

    The term ‘antimicrobial stewardship’ is defined as ‘an organisational or healthcare‑system‑wide
    approach to promoting and monitoring judicious use of antimicrobials to preserve
    their future effectiveness’.'
  - 'None.

    Date of next meeting: 1 December 2021 at 11am to 12.30pm'
- source_sentence: How much funding has the government committed to expand the Public
    Sector Fraud Authority to deploy AI in combating fraud?
  sentences:
  - '2) Embracing the opportunities presented by making greater use of cutting-edge
    technology, such as AI, across the public sector. The government is:

    More than doubling the size of i.AI, the AI incubator team, ensuring that the
    UK government has the in-house expertise consisting of the most talented technology
    professionals in the UK, who can apply their skills and expertise to appropriately
    seize the benefits of AI across the public sector and Civil Service.

    Committing £34 million to expand the Public Sector Fraud Authority by deploying
    AI to help combat fraud across the public sector, making it easier to spot, stop
    and catch fraudsters thereby saving £100 million for the public purse.

    Committing £17 million to accelerate DWP’s digital transformation, replacing paper-based
    processes with simplified online services, such as a new system for the Child
    Maintenance Service.

    Committing £14 million for public sector research and innovation infrastructure.
    This includes funding to develop the next generation of health and security technologies,
    unlocking productivity improvements in the public and private sector alike.

    3) Strengthening preventative action to reduce demand on public services. The
    government is:

    Committing an initial £105 million towards a wave of 15 new special free schools
    to create over 2,000 additional places for children with special educational needs
    and disabilities (SEND) across England. This will help more children receive a
    world-class education and builds on the significant levels of capital funding
    for SEND invested at the 2021 Spending Review. The locations of these special
    free schools will be announced by May 2024.

    Confirming the location of 20 Alternative Provision (AP) free schools, which will
    create over 1,600 additional AP places across England as part of the Spending
    Review 2021 commitment to invest £2.6 billion capital in high needs provision.
    This will support early intervention, helping improve outcomes for children requiring
    alternative provision, and helping them to fulfil their potential.'
  - "We will help build the UKDev (UK International Development) approach and brand\
    \ by leveraging the UK’s comparative advantage within both the public and private\
    \ sectors. We will build first and foremost on existing successful partnerships,\
    \ through which we share UK models and expertise to support digital transformation\
    \ in partner countries. For example, through our collaboration with the British\
    \ Standards Institution (BSI) we will expand our collaboration to build the capacity\
    \ of partner countries in Africa and South-East Asia (including through ASEAN)\
    \ on digital standards, working with local private sector and national standards-setting\
    \ bodies.\nWe will strengthen our delivery of peer learning activities in collaboration\
    \ with Ofcom, exchanging experiences and sharing the UK models on spectrum management,\
    \ local networks and other technical areas with telecoms regulators in partner\
    \ countries, building on the positive peer-learning experience with Kenya and\
    \ South Africa.\nWe will collaborate with Government Digital Service (GDS) to\
    \ share know-how with partner countries on digitalisation in the public sector,\
    \ building on our advisory role in GovStack[footnote 56]. We will leverage the\
    \ UK experience of DPI for public or regulated services (health, transport, banking,\
    \ land registries) based on the significant demand for this expertise from developing\
    \ countries and riding the momentum on DPI generated by the G20 India presidency\
    \ of 2023.\n 6.4 Enhancing FCDO’s digital development capability\nThe UK government\
    \ will also enhance its own digital development capability to keep up with the\
    \ pace of technological change, to be forward-looking and anticipate emergent\
    \ benefits and risks of digital transformation. We will invest in new research\
    \ on digital technologies and on their inclusive business models to build the\
    \ global evidence base, share lessons learned and improve knowledge management\
    \ through our portfolio of digital development and technology programmes, including\
    \ the FCDO’s new Technology Centre for Expertise (Tech CoE), which will complement\
    \ and support our programming portfolio.\nSince all sectors within international\
    \ development are underpinned by digital technologies, we will ensure that digital\
    \ development skills are mainstreamed across the FCDO. We will raise awareness\
    \ and upgrade staff knowledge through new training opportunities on best practice\
    \ in the complex and evolving area of digital development, through partnering\
    \ with existing FCDO capability initiatives, ie the International Academy’s Development\
    \ Faculty, the Cyber Network and the International Technology curriculum."
  - "The Burma (Sanctions) (EU Exit) Regulations 2019 (S.I. 2019/136) (revoked) 29\
    \ January 2019 To ensure that the UK continues to operate an effective sanctions\
    \ regime in relation to Burma after end of the Transition Period, replacing with\
    \ substantially the same effect the EU sanctions regime relating to Burma that\
    \ was previously in force in the UK under EU legislation and related UK legislation.\
    \ Section 2(4) report (PDF, 74 KB) and section 18 report (PDF, 65 KB).\nThe Burma\
    \ (Sanctions) (Overseas Territories) Order 2020 (S.I. 2020/1264) (revoked)[footnote\
    \ 81] 11 November 2020 To extend with modifications The Burma (Sanctions) (EU\
    \ Exit) Regulations 2019 (S.I. 2019/136) as amended from time to time to all British\
    \ Overseas Territories except Bermuda and Gibraltar (which implement sanctions\
    \ under their own legislative arrangements).  \nThe Myanmar (Sanctions) Regulations\
    \ 2021 (S.I. 2021/496) 26 April 2021 To establish a UK autonomous sanctions regime\
    \ in respect of Myanmar comprising financial, immigration and trade sanctions,\
    \ replacing the existing sanctions regime established by The Burma (Sanctions)\
    \ (EU Exit) Regulations 2019 (S.I. 2019/136).  \nThe Myanmar (Sanctions) (Overseas\
    \ Territories) Order 2021 (S.I. 2021/528) 28 April 2021 To extend with modifications\
    \ The Myanmar (Sanctions) Regulations 2021 (S.I. 2021/496) as amended from time\
    \ to time to all British Overseas Territories except Bermuda and Gibraltar (which\
    \ implement sanctions under their own legislative arrangements).  \nThe Myanmar\
    \ (Sanctions) (Isle of Man) Order 2021 (S.I. 2021/529) 28 April 2021 To extend\
    \ to the Isle of Man with modifications The Myanmar (Sanctions) Regulations 2021\
    \ (S.I. 2021/496) as amended from time to time.  \nSee also in section (C) of\
    \ this Annex:\nthe Sanctions Regulations (Commencement No. 1) (EU Exit) Regulations\
    \ 2019 (S.I. 2019/627)\nthe Sanctions (EU Exit) (Miscellaneous Amendments) (No.\
    \ 2) Regulations 2020 (S.I. 2020/590)\nthe Sanctions (EU Exit) (Miscellaneous\
    \ Amendments) (No. 4) Regulations 2020 (S.I. 2020/951)\nthe Sanctions (EU Exit)\
    \ (Miscellaneous Amendments) (No. 2) Regulations 2022 (S.I. 2022/818)\nStatutory\
    \ guidance for this regime was published on 29 April 2021.\n19. Nicaragua"
model-index:
- name: SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
  results:
  - task:
      type: semantic-similarity
      name: Semantic Similarity
    dataset:
      name: sts dev
      type: sts-dev
    metrics:
    - type: pearson_cosine
      value: 0.8601045098831278
      name: Pearson Cosine
    - type: spearman_cosine
      value: 0.8581596602965272
      name: Spearman Cosine
    - type: pearson_manhattan
      value: 0.8604789808039027
      name: Pearson Manhattan
    - type: spearman_manhattan
      value: 0.8571595448874573
      name: Spearman Manhattan
    - type: pearson_euclidean
      value: 0.8615938042335468
      name: Pearson Euclidean
    - type: spearman_euclidean
      value: 0.8581596602965272
      name: Spearman Euclidean
    - type: pearson_dot
      value: 0.8601045118561034
      name: Pearson Dot
    - type: spearman_dot
      value: 0.8581596602965272
      name: Spearman Dot
    - type: pearson_max
      value: 0.8615938042335468
      name: Pearson Max
    - type: spearman_max
      value: 0.8581596602965272
      name: Spearman Max
---

# SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a [sentence-transformers](https://www.SBERT.net) model finetuned from [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2). It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

It has been finetuned on a range of Q&A pairs based of [UK government policy documents.](https://huggingface.co/datasets/AndreasThinks/ukgov-policy-docs)

## Model Details

### Model Description
- **Model Type:** Sentence Transformer
- **Base model:** [sentence-transformers/all-MiniLM-L6-v2](https://huggingface.co/sentence-transformers/all-MiniLM-L6-v2) <!-- at revision 8b3219a92973c328a8e22fadcfa821b5dc75636a -->
- **Maximum Sequence Length:** 256 tokens
- **Output Dimensionality:** 384 tokens
- **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/UKPLab/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}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (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("AndreasThinks/all-MiniLM-L6-v2_policy_doc_finetune")
# Run inference
sentences = [
    'How much funding has the government committed to expand the Public Sector Fraud Authority to deploy AI in combating fraud?',
    '2) Embracing the opportunities presented by making greater use of cutting-edge technology, such as AI, across the public sector. The government is:\nMore than doubling the size of i.AI, the AI incubator team, ensuring that the UK government has the in-house expertise consisting of the most talented technology professionals in the UK, who can apply their skills and expertise to appropriately seize the benefits of AI across the public sector and Civil Service.\nCommitting £34 million to expand the Public Sector Fraud Authority by deploying AI to help combat fraud across the public sector, making it easier to spot, stop and catch fraudsters thereby saving £100 million for the public purse.\nCommitting £17 million to accelerate DWP’s digital transformation, replacing paper-based processes with simplified online services, such as a new system for the Child Maintenance Service.\nCommitting £14 million for public sector research and innovation infrastructure. This includes funding to develop the next generation of health and security technologies, unlocking productivity improvements in the public and private sector alike.\n3) Strengthening preventative action to reduce demand on public services. The government is:\nCommitting an initial £105 million towards a wave of 15 new special free schools to create over 2,000 additional places for children with special educational needs and disabilities (SEND) across England. This will help more children receive a world-class education and builds on the significant levels of capital funding for SEND invested at the 2021 Spending Review. The locations of these special free schools will be announced by May 2024.\nConfirming the location of 20 Alternative Provision (AP) free schools, which will create over 1,600 additional AP places across England as part of the Spending Review 2021 commitment to invest £2.6 billion capital in high needs provision. This will support early intervention, helping improve outcomes for children requiring alternative provision, and helping them to fulfil their potential.',
    'We will help build the UKDev (UK International Development) approach and brand by leveraging the UK’s comparative advantage within both the public and private sectors. We will build first and foremost on existing successful partnerships, through which we share UK models and expertise to support digital transformation in partner countries. For example, through our collaboration with the British Standards Institution (BSI) we will expand our collaboration to build the capacity of partner countries in Africa and South-East Asia (including through ASEAN) on digital standards, working with local private sector and national standards-setting bodies.\nWe will strengthen our delivery of peer learning activities in collaboration with Ofcom, exchanging experiences and sharing the UK models on spectrum management, local networks and other technical areas with telecoms regulators in partner countries, building on the positive peer-learning experience with Kenya and South Africa.\nWe will collaborate with Government Digital Service (GDS) to share know-how with partner countries on digitalisation in the public sector, building on our advisory role in GovStack[footnote 56]. We will leverage the UK experience of DPI for public or regulated services (health, transport, banking, land registries) based on the significant demand for this expertise from developing countries and riding the momentum on DPI generated by the G20 India presidency of 2023.\n 6.4 Enhancing FCDO’s digital development capability\nThe UK government will also enhance its own digital development capability to keep up with the pace of technological change, to be forward-looking and anticipate emergent benefits and risks of digital transformation. We will invest in new research on digital technologies and on their inclusive business models to build the global evidence base, share lessons learned and improve knowledge management through our portfolio of digital development and technology programmes, including the FCDO’s new Technology Centre for Expertise (Tech CoE), which will complement and support our programming portfolio.\nSince all sectors within international development are underpinned by digital technologies, we will ensure that digital development skills are mainstreamed across the FCDO. We will raise awareness and upgrade staff knowledge through new training opportunities on best practice in the complex and evolving area of digital development, through partnering with existing FCDO capability initiatives, ie the International Academy’s Development Faculty, the Cyber Network and the International Technology curriculum.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

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

<!--
### 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

#### Semantic Similarity
* Dataset: `sts-dev`
* Evaluated with [<code>EmbeddingSimilarityEvaluator</code>](https://sbert.net/docs/package_reference/sentence_transformer/evaluation.html#sentence_transformers.evaluation.EmbeddingSimilarityEvaluator)

| Metric              | Value      |
|:--------------------|:-----------|
| pearson_cosine      | 0.8601     |
| **spearman_cosine** | **0.8582** |
| pearson_manhattan   | 0.8605     |
| spearman_manhattan  | 0.8572     |
| pearson_euclidean   | 0.8616     |
| spearman_euclidean  | 0.8582     |
| pearson_dot         | 0.8601     |
| spearman_dot        | 0.8582     |
| pearson_max         | 0.8616     |
| spearman_max        | 0.8582     |

<!--
## 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 Hyperparameters
#### Non-Default Hyperparameters

- `eval_strategy`: steps
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `learning_rate`: 2e-05
- `num_train_epochs`: 2
- `warmup_ratio`: 0.1
- `use_mps_device`: True
- `batch_sampler`: no_duplicates

#### All Hyperparameters
<details><summary>Click to expand</summary>

- `overwrite_output_dir`: False
- `do_predict`: False
- `eval_strategy`: steps
- `prediction_loss_only`: True
- `per_device_train_batch_size`: 16
- `per_device_eval_batch_size`: 16
- `per_gpu_train_batch_size`: None
- `per_gpu_eval_batch_size`: None
- `gradient_accumulation_steps`: 1
- `eval_accumulation_steps`: None
- `learning_rate`: 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`: 2
- `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`: True
- `seed`: 42
- `data_seed`: None
- `jit_mode_eval`: False
- `use_ipex`: False
- `bf16`: False
- `fp16`: False
- `fp16_opt_level`: O1
- `half_precision_backend`: auto
- `bf16_full_eval`: False
- `fp16_full_eval`: False
- `tf32`: None
- `local_rank`: 0
- `ddp_backend`: None
- `tpu_num_cores`: None
- `tpu_metrics_debug`: False
- `debug`: []
- `dataloader_drop_last`: False
- `dataloader_num_workers`: 0
- `dataloader_prefetch_factor`: None
- `past_index`: -1
- `disable_tqdm`: False
- `remove_unused_columns`: True
- `label_names`: None
- `load_best_model_at_end`: False
- `ignore_data_skip`: False
- `fsdp`: []
- `fsdp_min_num_params`: 0
- `fsdp_config`: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
- `fsdp_transformer_layer_cls_to_wrap`: None
- `accelerator_config`: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
- `deepspeed`: None
- `label_smoothing_factor`: 0.0
- `optim`: adamw_torch
- `optim_args`: None
- `adafactor`: False
- `group_by_length`: False
- `length_column_name`: length
- `ddp_find_unused_parameters`: None
- `ddp_bucket_cap_mb`: None
- `ddp_broadcast_buffers`: False
- `dataloader_pin_memory`: True
- `dataloader_persistent_workers`: False
- `skip_memory_metrics`: True
- `use_legacy_prediction_loop`: False
- `push_to_hub`: False
- `resume_from_checkpoint`: None
- `hub_model_id`: None
- `hub_strategy`: every_save
- `hub_private_repo`: False
- `hub_always_push`: False
- `gradient_checkpointing`: False
- `gradient_checkpointing_kwargs`: None
- `include_inputs_for_metrics`: False
- `eval_do_concat_batches`: True
- `fp16_backend`: auto
- `push_to_hub_model_id`: None
- `push_to_hub_organization`: None
- `mp_parameters`: 
- `auto_find_batch_size`: False
- `full_determinism`: False
- `torchdynamo`: None
- `ray_scope`: last
- `ddp_timeout`: 1800
- `torch_compile`: False
- `torch_compile_backend`: None
- `torch_compile_mode`: None
- `dispatch_batches`: None
- `split_batches`: None
- `include_tokens_per_second`: False
- `include_num_input_tokens_seen`: False
- `neftune_noise_alpha`: None
- `optim_target_modules`: None
- `batch_eval_metrics`: False
- `batch_sampler`: no_duplicates
- `multi_dataset_batch_sampler`: proportional

</details>

### Training Logs
| Epoch  | Step | Training Loss | loss   | sts-dev_spearman_cosine |
|:------:|:----:|:-------------:|:------:|:-----------------------:|
| 0.0562 | 100  | 0.3598        | 0.8263 | 0.8672                  |
| 0.1124 | 200  | 0.1983        | 0.7948 | 0.8666                  |
| 0.1686 | 300  | 0.2021        | 0.7623 | 0.8666                  |
| 0.2248 | 400  | 0.1844        | 0.7510 | 0.8657                  |
| 0.2811 | 500  | 0.1704        | 0.7575 | 0.8629                  |
| 0.3373 | 600  | 0.1643        | 0.7348 | 0.8641                  |
| 0.3935 | 700  | 0.1808        | 0.7293 | 0.8604                  |
| 0.4497 | 800  | 0.1494        | 0.7232 | 0.8636                  |
| 0.5059 | 900  | 0.1563        | 0.7161 | 0.8634                  |
| 0.5621 | 1000 | 0.1345        | 0.7115 | 0.8643                  |
| 0.6183 | 1100 | 0.1344        | 0.7142 | 0.8617                  |
| 0.6745 | 1200 | 0.1584        | 0.7106 | 0.8622                  |
| 0.7307 | 1300 | 0.1488        | 0.7130 | 0.8592                  |
| 0.7870 | 1400 | 0.1391        | 0.7034 | 0.8635                  |
| 0.8432 | 1500 | 0.1433        | 0.7140 | 0.8614                  |
| 0.8994 | 1600 | 0.1393        | 0.7067 | 0.8612                  |
| 0.9556 | 1700 | 0.1644        | 0.6950 | 0.8628                  |
| 1.0118 | 1800 | 0.1399        | 0.7072 | 0.8594                  |
| 1.0680 | 1900 | 0.12          | 0.7093 | 0.8594                  |
| 1.1242 | 2000 | 0.0904        | 0.7040 | 0.8587                  |
| 1.1804 | 2100 | 0.082         | 0.6962 | 0.8585                  |
| 1.2366 | 2200 | 0.0715        | 0.6985 | 0.8593                  |
| 1.2929 | 2300 | 0.0624        | 0.7233 | 0.8562                  |
| 1.3491 | 2400 | 0.0725        | 0.7064 | 0.8581                  |
| 1.4053 | 2500 | 0.0665        | 0.7034 | 0.8570                  |
| 1.4615 | 2600 | 0.0616        | 0.6940 | 0.8584                  |
| 1.5177 | 2700 | 0.0703        | 0.6886 | 0.8599                  |
| 1.5739 | 2800 | 0.0564        | 0.6860 | 0.8603                  |
| 1.6301 | 2900 | 0.0603        | 0.6962 | 0.8590                  |
| 1.6863 | 3000 | 0.0729        | 0.6906 | 0.8589                  |
| 1.7426 | 3100 | 0.0753        | 0.6946 | 0.8579                  |
| 1.7988 | 3200 | 0.0711        | 0.6909 | 0.8582                  |
| 1.8550 | 3300 | 0.0743        | 0.6896 | 0.8583                  |
| 1.9112 | 3400 | 0.0693        | 0.6902 | 0.8581                  |
| 1.9674 | 3500 | 0.0845        | 0.6904 | 0.8582                  |


### Framework Versions
- Python: 3.10.13
- Sentence Transformers: 3.0.1
- Transformers: 4.41.2
- PyTorch: 2.3.1
- Accelerate: 0.31.0
- Datasets: 2.20.0
- Tokenizers: 0.19.1

## 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{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}
}
```

<!--
## 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.*
-->