Papers by Dorothea Kolossa
Who Wrote When? Author Diarization in Social Media Discussions (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches for author diarization are unable to detect stylistic shifts in a text . |
| Approach: | They propose a framework that integrates pre-trained neural representations of writing style with author-conditional encoder-decoder diarization. |
| Outcome: | The proposed framework is able to attribute comments in online discussions to individual authors. |
Variational Autoencoder with Embedded Student-t Mixture Model for Authorship Attribution (2020.coling-main)
Copied to clipboard
| Challenge: | Existing Variational autoencoders are limited by the assumed Gaussianity of the underlying probability distributions in the latent space. |
| Approach: | They propose a probabilistic autoencoding framework to deal with a supervised authorship attribution task. |
| Outcome: | The proposed method outperforms existing methods on an Amazon review dataset. |
MyFixit: An Annotated Dataset, Annotation Tool, and Baseline Methods for Information Extraction from Repair Manuals (2020.lrec-1)
Copied to clipboard
| Challenge: | Textual instructions are among the most widely used media for learning and teaching . however, to create autonomous systems, it is difficult to extract task knowledge from text . |
| Approach: | They propose methods that can extract information from repair manuals from a semi-structured dataset . they propose a bag-of-n-grams similarity method and deep-learning-based sequence labeling model . |
| Outcome: | The proposed methods can extract the needed tools and disassembled parts from repair manuals. |