Photo: alt left
Language Technology


Atom-feed of current news
UHH|4 April 2020

ACL 2020, the 58th Annual Meeting of the Association for Computational Linguistics, accepted the following long paper from the LT group:

Fynn Schröder and Chris Biemann: "Estimating the influence of auxiliary tasks for multi-task learning of sequence tagging tasks" (a pre-print is available here)
Verwaltung|25 March 2020

In this year's SemEval Shared Task 12 (OffensEval 2020) on "Multilingual Offensive Language Identification in Social Media", two systems developed by LT group members rank 1st and 2nd place out of 85 participating teams for Task A in the English language. The approach not only ranks high for Offensive Language Detection (A) but also for Offensive Language Categorization (Task B, 7th place) and...

UHH|17 December 2019
Our application for a collaboration project betwen literary science and language technology has been approved for funding by DFG. The project, conducted together with Evenyn Gius (TU Darmstadt), will use language processing for categorizing events and event structure in literary texts, and is associated to the DFG-SPP 2207 "Computational Literary Studies".
UHH|30 November 2019
We happily announce that the second phase of the 2nd phase of Transregio-SFB "Crossmodal Learning" (Hamburg - Beijing) has been approved for funding by DFG. In our new sub-project C7 "Crossmodal learning for improving human reading", we will utilize crossmodal embeddings for understanding and facilitating the process of learning how to read in collaboration with psychologists from the Chinese...
UHH|14 October 2019

Every two years the GSCL (German Society of Computational Linguists) awards the best Bachelor and Master thesis. Three finalists were invited to present their thesis at the German conference for processing natural language: KONVENS 2019. This year's nominations included Rami Aly for his thesis 'Hierarchical Genre Classification using Neural Networks' supervised by Steffen Remus and Chris Biemann.


UHH|1 October 2019

A new publication from LT group member Gregor Wiedemann has appeared in "Language Resources and Evaluation":

Wiedemann, G.; Heyer, G. (2019): Multi-modal page stream segmentation with convolutional neural networks, In: Language Resources and Evaluation (LREV), Online first: 27.09.2019.

The paper introduces an approach to combine image and text information for document flow separation and evaluates...

UHH|1 September 2019
Dr. Gregor Wiedemann from the Language Technology group of our department has successfully acquired a new DFG project. In the project "A Framework for Argument Mining and Evaluation (FAME)", he will investigate together with researchers from the computer science department at Leipzig University (Dr. Ringo Baumann and Prof. Gerhard Heyer) how argument mining in the area of natural language...
UHH|31 July 2019
We are happy to announce the release of version 2.0 of our software new/s/leak for information extraction and interactive visualization of textual datasets for investigative journalism and eDiscovery.
UHH|8 July 2019

Our new article on local-global graphclustering with Watset has just been accepted for the September issue of the Computational Linguistics journal:

Dmitry Ustalov, Alexander Panchenko, Chris Biemann and Simone Paolo Ponzetto (2019): Watset: Local-Global Graph Clustering with Applications in Sense and Frame Induction. Posted Online June 25, 2019,  (link)


Verwaltung|3 July 2019

The 20th Conference of the International Speech Communication Association (INTERSPEECH2019) accepted the following paper from our lab :

Benjamin Milde and Chris Biemann, "SparseSpeech: Unsupervised Acoustic Unit Discovery with Memory-Augmented Sequence Autoencoders "

A preprint version of the paper is available here