The researcher will be supervised by Prof. Andreas Geiger (University of Tübingen) and Iryna Gurevych (TU Darmstadt) and will have the opportunity to supervise Master and PhD students.
The body of scientific literature is growing at an ever-increasing rate. As a result, it is increasingly difficult for researchers to keep up-to-date. This hinders scientific progress at large and leads to a suboptimal usage of resources including research funds, compute, energy and intellectual capacity. In this project, we plan to develop novel NLP methods and algorithms and to collect new datasets to advance research in scientific documents processing. Research topics include:
- Efficient hierarchical and multi-modal document representations
- Structured intra- and inter-document models
- Distillation and adaptation of LLMs for scientific document analysis and generation
- Self-supervised learning with multi-scale pre-text tasks
- Explainable and grounded scientific document models
- Deployment of algorithms and collection of datasets (www.scholar-inbox.com)
We are looking for candidates that hold a PhD degree and who have published at top conferences in the field (ACL, EMNLP, NAACL, TACL).
The University of Tübingen is one of Germany’s excellence universities with an excellence cluster on machine learning, an ELLIS Unit and the Tübingen AI Center. Embedded in the interdisciplinary research environment of CyberValley, the Autonomous Vision Group conducts curiosity-driven fundamental research, providing researchers with access to unique research facilities and great research teams. Currently, 2 PhD students are working on this project. Our culture is international, inclusive and collaborative. We are looking forward to your application!
To apply, please send your application materials including your CV, research statement, transcripts and names of referees to: email@example.com