This workshop is directly associated to the NeurIPS 2023 competition on “Privacy Preserving Federated Learning Document VQA” (PFL-DocVQA).
The objective of PFL-DocVQA is to develop privacy-preserving solutions for fine-tuning multi-modal language models for document understanding on distributed data. We seek efficient federated learning solutions for fine-tuning a pre-trained generic Document Visual Question Answering (DocVQA) model on a new domain, that of invoice processing.
Automatically managing the information of document workflows is a core aspect of business intelligence and process automation. Reasoning over the information extracted from documents fuels subsequent decision-making processes that can directly affect humans, especially in sectors such as finance, legal or insurance. At the same time, documents tend to contain private information, restricting access to them during training. This common scenario requires training large-scale models over private and widely distributed data.
This workshop aims to highlight the important synergies between the Document Intelligence and the Privacy communities. During the workshop, we will review the NeurIPS competition setup and the results of submitted methods, while we will invite winning teams from the competition to present their methods.
This will be coupled with invited talks on differential privacy, federated learning and document intelligence.
The PFL-DocVQA competition and associated workshop are organised under the aegis of the European Network on Safe and Secure AI (ELSA).
