Search Results for author: Lars Hillebrand

Found 7 papers, 3 papers with code

Pointer-Guided Pre-Training: Infusing Large Language Models with Paragraph-Level Contextual Awareness

1 code implementation6 Jun 2024 Lars Hillebrand, Prabhupad Pradhan, Christian Bauckhage, Rafet Sifa

We introduce "pointer-guided segment ordering" (SO), a novel pre-training technique aimed at enhancing the contextual understanding of paragraph-level text representations in large language models.

Diversity text-classification +1

sustain.AI: a Recommender System to analyze Sustainability Reports

1 code implementation15 May 2023 Lars Hillebrand, Maren Pielka, David Leonhard, Tobias Deußer, Tim Dilmaghani, Bernd Kliem, Rüdiger Loitz, Milad Morad, Christian Temath, Thiago Bell, Robin Stenzel, Rafet Sifa

We present sustainAI, an intelligent, context-aware recommender system that assists auditors and financial investors as well as the general public to efficiently analyze companies' sustainability reports.

Multi-Label Classification Recommendation Systems

Towards automating Numerical Consistency Checks in Financial Reports

no code implementations11 Nov 2022 Lars Hillebrand, Tobias Deußer, Tim Dilmaghani, Bernd Kliem, Rüdiger Loitz, Christian Bauckhage, Rafet Sifa

It combines a financial named entity and relation extraction module with a BERT-based filtering and text pair classification component to extract KPIs from unstructured sentences before linking them to synonymous occurrences in the balance sheet and profit & loss statement.

Relation Extraction Text Pair Classification

KPI-EDGAR: A Novel Dataset and Accompanying Metric for Relation Extraction from Financial Documents

1 code implementation17 Oct 2022 Tobias Deußer, Syed Musharraf Ali, Lars Hillebrand, Desiana Nurchalifah, Basil Jacob, Christian Bauckhage, Rafet Sifa

We introduce KPI-EDGAR, a novel dataset for Joint Named Entity Recognition and Relation Extraction building on financial reports uploaded to the Electronic Data Gathering, Analysis, and Retrieval (EDGAR) system, where the main objective is to extract Key Performance Indicators (KPIs) from financial documents and link them to their numerical values and other attributes.

Benchmarking Joint Entity and Relation Extraction +5

KPI-BERT: A Joint Named Entity Recognition and Relation Extraction Model for Financial Reports

no code implementations3 Aug 2022 Lars Hillebrand, Tobias Deußer, Tim Dilmaghani, Bernd Kliem, Rüdiger Loitz, Christian Bauckhage, Rafet Sifa

We present KPI-BERT, a system which employs novel methods of named entity recognition (NER) and relation extraction (RE) to extract and link key performance indicators (KPIs), e. g. "revenue" or "interest expenses", of companies from real-world German financial documents.

named-entity-recognition Named Entity Recognition +4

Cannot find the paper you are looking for? You can Submit a new open access paper.