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.
2082k
1Geographical allocation of invested plan assets231 Dec 2023 31 Dec 20223in € m. (unless stated otherwise) Germany4United5Kingdom6United7States8Other 9Eurozone10Other 11developed 12countries13Emerging14markets Total Germany15United16Kingdom17United18States19Other 20Eurozone21Other 22developed 23countries24Emerging25markets Total26Cash and cash equivalents 0 0 1 10 2 1 14 0 0 2 2 1 2 727Equity instruments 1 2 36 13 9 2 63 1 1 27 12 6 1 4928Government bonds29(investment-grade and above) 16 0 3 35 9 19 82 13 0 3 39 4 15 7530Government bonds31(non-investment-grade) 0 0 0 0 0 4 4 0 0 0 0 0 4 432Non-government bonds33(investment-grade and above)15 4 56 71 18 1 166 10 9 50 65 15 1 15034Non-government bonds35(non-investment-grade) 0 0 1 13 0 0 14 0 0 0 11 0 0 1236Securitised and other debt 37investments 0 0 0 0 1 0 1 0 0 0 0 0 0 138Subtotal 32 7 96 144 39 27 344 25 10 82 129 28 23 29639Share (in %) 9 2 28 42 11 8 100 8 3 28 44 9 8 10040Other asset categories 114 12241Fair value of plan assets 458 41942Plan assets included derivative transactions with other Deutsche Bank Group entities with a 43market value of positive € 33 million and positive € 46 million at 31 December 2023 and 4431 December 2022, respectively. There were neither a material number of securities issued by 45the Group nor other claims against the Group assets included in the fair value of plan assets. 46The plan assets did not include any real estate which is used by the Group.47Key Risk Sensitivities48The Group’s defined benefit obligations are sensitive to changes in capital market conditions 49and actuarial assumptions. Sensitivities to capital market movements and key assumption 50changes are presented in the following table. Each market risk factor or assumption is 51changed in isolation. Sensitivities of the defined benefit obligations are approximated using 52geometric extrapolation methods based on plan durations for the respective assumption. 53Duration is a risk measure that indicates the broad sensitivity of the obligations to a change in 54an underlying assumption and provides a reasonable approximation for small to moderate 55changes in those assumptions.56For example, the interest rate duration is derived from the change in the defined benefit 57obligation to a change in the interest rate based on information provided by the local 58actuaries of the respective plans. The resulting duration is used to estimate the 59remeasurement liability loss or gain from changes in the interest rate. For other assumptions, 60a similar approach is used to derive the respective sensitivity results.61For defined benefit pension plans, changes in capital market conditions will impact the plan 62obligations via actuarial assumptions – mainly interest rate and inflation rate – as well as the 63plan assets’ fair value. Where the Group applies a liability driven investment approach, the 64overall exposure to such changes is reduced. To help readers gain a better understanding of 65the Group’s risk exposures to key capital market movements, the net impact of the change in 66the defined benefit obligations and plan assets due to a change of the related market risk 67factor or underlying actuarial assumption is shown. Where changes in actuarial assumptions 68do not affect the plan assets, only the impact on the defined benefit obligations is reported. 69Asset-related sensitivities are derived for major plans which are applicable to the Group by 70using risk sensitivity factors determined by Deutsche Bank Group’s market risk management 71function. These sensitivities are calculated based on information provided by the plans’ 72investment managers and extrapolated linearly to reflect the approximate change of the plan 73assets’ market value in case of a change in the underlying risk factor.74 75To our 76Shareholders77Summarised 78Management Report79Consolidated 80Financial Statements Compensation Report Corporate Govern-81ance Statement82Supplementary 83Information DWS 2023 Annual Report84 85Additional Notes86120 19 – Employee Benefits