AmazonScience/document-haystack
Document Haystack Dataset This repository contains the dataset for the paper “Document Haystack: A Long Context Multimodal Image/Document Understanding Vision LLM Benchmark”. 📑 Abstract Paper The proliferation of multimodal Large Language Models has significantly advanced the ability to analyze and understand complex data inputs from different modalities. However, the processing of long documents remains under-explored, largely due to a lack of suitable… See the full description on the dataset page: https://huggingface.co/datasets/AmazonScience/document-haystack.
2090k
1Financed emissions continued2Setting our targets 3Our target-setting approach to date, for 4on-balance sheet financed emissions and 5facilitated emissions, has been to utilise a 6single net zero reference scenario (IEA NZE 72021) to underpin both energy supply-related 8sectors (oil and gas, power and utilities, 9and thermal coal mining) and our published 10targets for demand-side sectors in transport 11and heavy industry. 12The impact of our capital markets activities 13is now reflected in our combined financed 14emissions targets for the oil and gas, and 15power and utilities sectors. Our facilitated 16emissions, included in our combined metrics, 17are weighted at 33%, in accordance with 18the PCAF standard. This approach dampens 19volatility, apportions responsibility between 20underwriters and asset owners, and allows 21for flexibility in deploying on and off-balance 22sheet financing in line with clients’ needs. 23To further reduce the inherent volatility in 24facilitated emissions, we apply a three-year 25moving average across transactions for our 26target metric, building up from 2019 data. This 27means that transactions facilitated in 2028 28and 2029 will still have an impact on the 2030 29progress number and will need to be taken 30into consideration as we manage progress 31towards our target. We aim to achieve our 32target in 2030 notwithstanding the application 33of a three-year average. 34Our approach for financed emissions 35accounting does not rely on purchasing 36offsets to achieve any financed emissions 37targets we set.38An evolving approach39We believe methodologies for calculating 40financed emissions and setting targets should 41be transparent and comparable, and should 42provide science-based insights that focus 43engagement efforts, inform capital allocation 44and support the development of solutions that 45are both timely and impactful. We continue to 46engage with regulators, standard setters and 47industry bodies to help shape our approach to 48measuring financed emissions and managing 49portfolio alignment to net zero. We also work 50with data providers and our clients to help us 51gather data from the real economy to improve 52our analysis. 53Scenarios used in our analysis are modelled 54on assumptions of the available carbon 55budget and actions that need to be taken to 56limit the long-term increase in average global 57temperatures to 1.5°C with limited overshoot. 58We expect that the scenarios we use will 59be updated periodically. We plan to refine 60our own analysis of financed emissions as 61industry guidance on scenarios, data and 62methodologies more broadly evolve in the 63years ahead.64Agriculture 65For the agriculture sector, due to ongoing 66data availability and quality challenges, and 67lack of developed methodologies, we are not 68in a position to report our financed emissions 69or set a target at this time. We aim to build 70data availability and continue to work with 71partners and industry bodies to develop data 72and methodologies across a wider section of 73the agriculture value chain – such as farm-74related and downstream emissions, including 75from the food and beverage sector – while 76assessing the make-up of our portfolio.77Residential real estate 78For residential real estate, where our 79customers are consumers not corporates, 80our approach needs to consider financial 81inclusivity, and our ability to provide customers 82access to suitable mortgages in addition to 83decarbonisation aims. We expect to measure 84and report our residential real estate financed 85emissions in future disclosures. We continue 86to consider our approach to setting an 87appropriate target to measure our contribution 88to helping the sector transition.89Commercial real estate 90For commercial real estate, we continue to 91work towards outlining a baseline and a 2030 92financed emissions ambition or ambition 93range, starting with our major markets and 94where sufficient data is available to track 95decarbonisation progress. We expect to 96review our approach and coverage periodically 97in line with evolving data, methodologies, 98scenarios and real-world progress. 99Methodologies for embedded carbon 100need to be developed given the materiality 101of financing new property development 102within our portfolio, from a financed 103emissions perspective.104Investing in battery 105health and monitoring 106solutions 107The global push towards electrification 108is accelerating the demand for 109systems powered by safe, reliable and 110sustainable batteries. 111In August 2023, HSBC Asset 112Management, as part of its climate 113tech venture capital strategy, helped a 114Germany-based analytics software start-115up secure $7.8m (€7.2m) of investment 116in its battery monitoring platform, 117with HSBC Asset Management’s fund 118providing $4.1m (€3.8m).119ACCURE Battery Intelligence uses AI, 120field data and modelling to forecast and 121manage the health and performance 122of batteries, and predict failures, fires 123and other incidents. With their software 124already supporting 3.5 gigawatt-hours 125of storage, the fundraising will help 126expand and develop the platform 127across energy, electric vehicle, transit, 128marine, insurance and other industries 129worldwide.13054 HSBC Holdings plc Annual Report and Accounts 2023131ESG review | Environmental 