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.
2089k
1assets under management from our collective and individual portfolio management activities 2without investments in central governments, central banks and supranational issuers. For the 3avoidance of doubt, references to “Group AuM” in this section are to AuM of DWS Group 4calculated in accordance with the foregoing FAQ. 5Further, we have not used estimates for assessing Taxonomy-alignment of our investments.6We used data from external data service providers for our Article 8 Taxonomy Regulation 7reporting. For our Illiquid businesses, data sources used include internal sources and external 8counterparties such as investee companies and fund managers.9Further, based on guidance entitled “Frequently asked questions: How should financial and 10non-financial undertakings report taxonomy-eligible economic activities and assets in 11accordance with the Taxonomy Regulation Article 8 Disclosures Delegated Act?” by the 12European Commission on 31 January 2022, we applied a look-through approach to the 13majority of our fund holdings, to the extent relevant data was available, to improve our 14reporting.15ESG data on all of our directly managed Real Estate assets is incorporated in a third-party ESG 16data management platform (Measurabl). Taxonomy-alignment for Real Estate assets is to be 17assessed applying the technical screening criteria of the Delegated Regulation (EU) 182021/2139.19Real Estate assets outside of the EU, with the exception of those which have a UK Energy 20Performance Certificate, are not subject to the EU Energy Performance Certificates regime 21nor the EU Energy Performance of Buildings Directive, which are required for technical 22screening for the environmental objective “Climate Change Mitigation” of the Taxonomy 23Regulation. At present, there is no established and widely recognised international method 24available to translate various global non-EU energy performance assessment schemes into EU 25Energy Performance Certificates. Therefore, we believe it is currently not possible to 26accurately assess climate mitigation performance of non-EU/non-UK Real Estate assets.27We applied the substantial contribution technical screening criteria for Climate Change 28Mitigation according to the Delegated Regulation (EU) 2021/2139 to our directly managed 29Real Estate assets located in the EU and the UK. The result of this analysis was that only an 30immaterial percentage fulfilled those criteria in 2023. 31For all of our directly managed Real Estate assets we could not completely assess Taxonomy-32alignment according to the technical screening criteria of Delegated Regulation (EU) 332021/2139 with the environmental objective of Climate Change Adaptation of the Taxonomy 34Regulation because we believe it is at present not possible to accurately assess climate 35change adaptation due to inadequate methodology, capacity and verification systems36As a consequence, the weighted average value of all investments that are directed at funding, 37or are associated with taxonomy-aligned economic activities used for calculating our KPI does 38not include the value of our directly managed Real Estate assets.39ESG data on all our indirectly managed Real Estate assets is not available in Measurabl, 40because we do not have sufficient access to these assets at this point in time. Therefore, due 41to lack of data availability, it is not possible to assess Taxonomy-alignment for these assets. 42Indirectly managed assets are assets the value of which is included in our collective and/or 43individual portfolio management activities as we are ultimately responsible for their 44management, but which are directly managed by a third party.45All data have limitations which include the reliance on external valuation methodology, data 46availability and data quality such as completeness and correctness that can result in over- or 47understating of the KPI. The information is provided as of 31 December 2023. 48Compliance of our Business Strategy, Product Design Processes and 49Engagement with Clients and Counterparties50Notwithstanding the many challenges in obtaining reliable data for the economic activities 51covered by the Taxonomy Regulation, we will continue to seek to comply with it in relation to 52our business strategy, product design processes and engagement with clients and 53counterparties.54We also aim to make further use of data reported in accordance with the Taxonomy 55Regulation in our research activities as well as in proxy voting and engagement, where such 56data are available and reliable.57 58To our Shareholders Summarised 59Management Report60Consolidated 61Financial Statements Compensation Report Corporate Govern-62ance Statement63Supplementary 64Information DWS 2023 Annual Report65 66Complementary Information6768 Disclosures in Accordance with Article 8 Taxonomy Regulation and Delegated Regulation (EU) 2021/2178