Climate Check Reranker

🚀 Climate-Science-Reranker

This is a Cross Encoder model that computes scores for text pairs, useful for text reranking and semantic search.

✨ Features

• It's a Cross Encoder model finetuned from cross-encoder/ms-marco-MiniLM-L6-v2 using the sentence-transformers library.
• It can compute scores for pairs of texts, which can be applied to text reranking and semantic search.

📚 Documentation

Model Details

Model Description

Model Sources

• Documentation: Sentence Transformers Documentation
• Documentation: Cross Encoder Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Cross Encoders on Hugging Face

💻 Usage Examples

Basic Usage

pip install -U sentence-transformers

from sentence_transformers import CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("cross_encoder_model_id")
# Get scores for pairs of texts
pairs = [
    ['Arctic sea ice is shrinking, while Antarctic ice is growing. #ClimateScience #PolarRegions', 'Abstract The adoption of healthy diets with low environmental impact has been widely promoted as an important climate change mitigation strategy. Typically, these diets are high in plant-sourced and low in animal-sourced and processed foods. Despite the fact that their environmental impacts vary, they are often referred to as ‘sustainable diets’. Here we systematically review the available published evidence on the effect of ‘sustainable diets’ on environmental footprints and human health. Eight databases (OvidSP-Medline, OvidSP-Embase, EBSCO-GreenFILE, Web of Science Core Collection, Scopus, OvidSP-CAB-Abstracts, OvidSP-AGRIS, and OvidSP-Global Health) were searched to identify literature (published 1999–2019) reporting health effects and environmental footprints of ‘sustainable diets’. Available evidence was mapped and pooled analysis was conducted by unique combinations of diet pattern, health and environmental outcome. Eighteen studies (412 measurements) met our inclusion criteria, distinguishing twelve non-mutually exclusive sustainable diet patterns, six environmental outcomes, and seven health outcomes. In 87% of measurements (n = 151) positive health outcomes were reported from ‘sustainable diets’ (average relative health improvement: 4.09% [95% CI −0.10–8.29]) when comparing ‘sustainable diets’ to current/baseline consumption patterns. Greenhouse gas emissions associated with ‘sustainable diets’ were on average 25.8%[95%CI −27.0 to −14.6] lower than current/baseline consumption patterns, with vegan diets reporting the largest reduction in GHG-emissions (−70.3% [95% CI: −90.2 to −50.4]), however, water use was frequently reported to be higher than current/baseline diets. Multiple benefits for both health and the environment were reported in the majority (n = 315[76%]) of measurements. We identified consistent evidence of both positive health effects and reduced environmental footprints accruing from ‘sustainable diets’. The notable exception of increased water use associated with ‘sustainable diets’ identifies that co-benefits are not universal and some trade-offs are likely. When carefully designed, evidence-based, and adapted to contextual factors, dietary change could play a pivotal role in climate change mitigation, sustainable food systems, and future population health.'],
    ['We can accelerate our transition away from fossil fuels in order to mitigate future warming.', 'The recent debate on the temporal dynamics of energy transitions is crucial since one of the main reasons for embarking on transitions away from fossil fuels is tackling climate change. Long-drawn out transitions, taking decades or even centuries as we have seen historically, are unlikely to help achieve climate change mitigation targets. Therefore, the pace of energy transitions and whether they can be sped up is a key academic and policy question. Our argument is that while history is important in order to understand the dynamics of transitions, the pace of historic transitions is only partly a good guide to the future. We agree with Sovacool’s [1] argument that quicker transitions have happened in the past and may therefore also be possible in the future globally. The key reason for our optimism is that historic energy transitions have not been consciously governed, whereas today a wide variety of actors is engaged in active attempts to govern the transition towards low carbon energy systems. In addition, international innovation dynamics can work in favour of speeding up the global low-carbon transition. Finally, the 2015 Paris agreement demonstrates a global commitment to move towards a low carbon economy for the first time, thereby signalling the required political will to foster quick transitions and to overcome resistance, such as from incumbents with sunk infrastructure investments.'],
    ["We've known about climate change for a long time. #ClimateActionNow", 'The recent warming in the Arctic is affecting a broad spectrum of physical, ecological, and human/cultural systems that may be irreversible on century time scales and have the potential to cause rapid changes in the earth system. The response of the carbon cycle of the Arctic to changes in climate is a major issue of global concern, yet there has not been a comprehensive review of the status of the contemporary carbon cycle of the Arctic and its response to climate change. This review is designed to clarify key uncertainties and vulnerabilities in the response of the carbon cycle of the Arctic to ongoing climatic change. While it is clear that there are substantial stocks of carbon in the Arctic, there are also significant uncertainties associated with the magnitude of organic matter stocks contained in permafrost and the storage of methane hydrates beneath both subterranean and submerged permafrost of the Arctic. In the context of the global carbon cycle, this review demonstrates that the Arctic plays an important role in the global dynamics of both CO2 and CH4. Studies suggest that the Arctic has been a sink for atmospheric CO2 of between 0 and 0.8 Pg C/yr in recent decades, which is between 0% and 25% of the global net land/ocean flux during the 1990s. The Arctic is a substantial source of CH4 to the atmosphere (between 32 and 112 Tg CH4/yr), primarily because of the large area of wetlands throughout the region. Analyses to date indicate that the sensitivity of the carbon cycle of the Arctic during the remainder of the 21st century is highly uncertain. To improve the capability to assess the sensitivity of the carbon cycle of the Arctic to projected climate change, we recommend that (1) integrated regional studies be conducted to link observations of carbon dynamics to the processes that are likely to influence those dynamics, and (2) the understanding gained from these integrated studies be incorporated into both uncoupled and fully coupled carbon–climate modeling efforts.'],
    ['Fruit trees produce fresh oxygen.', 'Oxygen has shaped life on Earth as we know it today. Molecular oxygen is essential for normal cellular function, i.e., plants need oxygen to maintain cellular respiration and for a wide variety of biochemical reactions. When oxygen levels in the cell are lower than levels needed for respiration, then the cell experiences hypoxia. Plants are known to experience root hypoxia during natural environmental conditions like flooding. Fruit, on the other hand, is known to be hypoxic under normal oxygen conditions. This observation could be explained (at least partially) as a consequence of diffusional barriers, low tissue diffusivity, and high oxygen consumption by respiration. From the physiological point of view, hypoxia is known to have a profound impact on fruit development, since it is well documented that a low oxygen environment can significantly delay ripening and senescence of some fruit. This effect of a low-oxygen environment is readily used for optimizing storage conditions and transport, and for prolonging the shelf life of several fruit commodities. Therefore, further understanding of the complex relationship between oxygen availability within the cell and fruit development could assist postharvest management.'],
    ["Did you know that when an iceberg melts, it doesn't actually raise sea levels?  🤯  It's because the ice was already displacing water when it was a berg. #sciencefacts #oceanfacts", 'Abstract. The drastic reduction of the Arctic sea ice over the past 40 years is the most glaring evidence of climate change on Planet Earth. Among all the variables characterizing sea ice, the sea ice volume is by far the most sensitive one for climate change since it is decaying at the highest rate compared to sea ice extent and sea ice thickness. In 40 years the Arctic Ocean has lost about 3/4 of its sea ice volume at the end of the summer season corresponding to a reduction of both sea ice extent and sea ice thickness by half on average. From more than 16 000 km3, 40 years ago, the Arctic sea ice summer minimum dropped down to less than 4000 km3 during the most recent summers. Being a combination of Arctic sea ice extent and sea ice thickness, the Arctic sea ice volume is difficult to observe directly and accurately. We estimated cumulative Freezing-Degree Days (FDD) over a 9 month freezing time period (September to May each year) based on ERA Interim surface air temperature reanalysis over the whole Arctic Ocean and for the past 38 years. Then we compared the Arctic sea ice volume based on sea ice thickness deduced from cumulative FDD with Arctic sea ice volume estimated from PIOMAS (Pan Arctic Ice Ocean Modeling and Assimilation System) and from the ESA CRYOSAT-2 satellite. The results are strikingly similar. The warming of the atmosphere is playing an important role in contributing to the Arctic sea ice volume decrease during the whole freezing season (September to May). In addition, the FDD spatial distribution exhibiting a sharp double peak-like feature is reflecting the Multi Y ear Ice (MYI) versus First Year Ice (FYI) dual disposition typical of the Arctic sea ice cover. This is indicative of a significant contribution from the vertical ocean heat fluxes throughout the ice depending on MYI versus FYI distribution and the snow layer on top of it influencing the surface air temperature accordingly. In 2018 the Arctic MYI vanished almost completely for the first time ever over the past 40 years. The quasi complete disappearance of the Arctic sea ice is more likely to happen in summer within the next 15 years with broad consequences for Arctic marine and terrestrial ecosystems, climate and weather patterns on a planetary scale and globally on human activities.\n'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

Advanced Usage

# Rank different texts based on similarity to a single text
ranks = model.rank(
    'Arctic sea ice is shrinking, while Antarctic ice is growing. #ClimateScience #PolarRegions',
    [
        'Abstract The adoption of healthy diets with low environmental impact has been widely promoted as an important climate change mitigation strategy. Typically, these diets are high in plant-sourced and low in animal-sourced and processed foods. Despite the fact that their environmental impacts vary, they are often referred to as ‘sustainable diets’. Here we systematically review the available published evidence on the effect of ‘sustainable diets’ on environmental footprints and human health. Eight databases (OvidSP-Medline, OvidSP-Embase, EBSCO-GreenFILE, Web of Science Core Collection, Scopus, OvidSP-CAB-Abstracts, OvidSP-AGRIS, and OvidSP-Global Health) were searched to identify literature (published 1999–2019) reporting health effects and environmental footprints of ‘sustainable diets’. Available evidence was mapped and pooled analysis was conducted by unique combinations of diet pattern, health and environmental outcome. Eighteen studies (412 measurements) met our inclusion criteria, distinguishing twelve non-mutually exclusive sustainable diet patterns, six environmental outcomes, and seven health outcomes. In 87% of measurements (n = 151) positive health outcomes were reported from ‘sustainable diets’ (average relative health improvement: 4.09% [95% CI −0.10–8.29]) when comparing ‘sustainable diets’ to current/baseline consumption patterns. Greenhouse gas emissions associated with ‘sustainable diets’ were on average 25.8%[95%CI −27.0 to −14.6] lower than current/baseline consumption patterns, with vegan diets reporting the largest reduction in GHG-emissions (−70.3% [95% CI: −90.2 to −50.4]), however, water use was frequently reported to be higher than current/baseline diets. Multiple benefits for both health and the environment were reported in the majority (n = 315[76%]) of measurements. We identified consistent evidence of both positive health effects and reduced environmental footprints accruing from ‘sustainable diets’. The notable exception of increased water use associated with ‘sustainable diets’ identifies that co-benefits are not universal and some trade-offs are likely. When carefully designed, evidence-based, and adapted to contextual factors, dietary change could play a pivotal role in climate change mitigation, sustainable food systems, and future population health.'
    ]
)

📄 License

This model is released under the apache-2.0 license.