Search Results for author: Felix A. Gers

Found 10 papers, 8 papers with code

VisBERT: Hidden-State Visualizations for Transformers

1 code implementation9 Nov 2020 Betty van Aken, Benjamin Winter, Alexander Löser, Felix A. Gers

At the same time, they are difficult to incorporate into the large, black-box models that achieve state-of-the-art results in a multitude of NLP tasks.

Multi-hop Question Answering Question Answering

Learning Contextualized Document Representations for Healthcare Answer Retrieval

1 code implementation3 Feb 2020 Sebastian Arnold, Betty van Aken, Paul Grundmann, Felix A. Gers, Alexander Löser

Our model leverages a dual encoder architecture with hierarchical LSTM layers and multi-task training to encode the position of clinical entities and aspects alongside the document discourse.

Passage Ranking Retrieval +1

How Does BERT Answer Questions? A Layer-Wise Analysis of Transformer Representations

2 code implementations11 Sep 2019 Betty van Aken, Benjamin Winter, Alexander Löser, Felix A. Gers

In order to better understand BERT and other Transformer-based models, we present a layer-wise analysis of BERT's hidden states.

Question Answering

SECTOR: A Neural Model for Coherent Topic Segmentation and Classification

3 code implementations TACL 2019 Sebastian Arnold, Rudolf Schneider, Philippe Cudré-Mauroux, Felix A. Gers, Alexander Löser

From our extensive evaluation of 20 architectures, we report a highest score of 71. 6% F1 for the segmentation and classification of 30 topics from the English city domain, scored by our SECTOR LSTM model with bloom filter embeddings and bidirectional segmentation.

Classification General Classification +2

IDEL: In-Database Entity Linking with Neural Embeddings

no code implementations13 Mar 2018 Torsten Kilias, Alexander Löser, Felix A. Gers, Richard Koopmanschap, Ying Zhang, Martin Kersten

We present a novel architecture, In-Database Entity Linking (IDEL), in which we integrate the analytics-optimized RDBMS MonetDB with neural text mining abilities.

Entity Linking Retrieval

Analysing Errors of Open Information Extraction Systems

no code implementations WS 2017 Rudolf Schneider, Tom Oberhauser, Tobias Klatt, Felix A. Gers, Alexander Löser

We report results on benchmarking Open Information Extraction (OIE) systems using RelVis, a toolkit for benchmarking Open Information Extraction systems.

Benchmarking Open Information Extraction

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