Robert Litschko

Robert Litschko

Ph.D. candidate

University of Mannheim

Research

I’m a computer science PhD student at the Data and Web Science Group at the University of Mannheim, where I am supervised by Prof. Goran Glavaš. My interests lie in transfer learning for Natural Language Processing (NLP) and Information Retrieval (IR). Currently, I am very interested in resource-lean transfer methods including, but not limited to, unsupervised, few-shot, and zero-shot learning applied to cross-lingual IR and NLP.

Interests

  • Natural Language Processing
  • Information Retrieval
  • Cross-lingual Transfer Learning
  • Representation Learning

Education

  • PhD in Computer Science

    University of Mannheim

  • MSc (hon.) in Business Informatics, 2017

    University of Mannheim

  • BSc in Business Informatics, 2014

    Karlsruhe University of Applied Sciences

Publications

(2021). Evaluating Multilingual Text Encoders for Unsupervised Cross-Lingual Retrieval. Proceedings of ECIR.

(2020). Probing Pretrained Language Models for Lexical Semantics. Proceedings of EMNLP.

PDF Video

(2020). Towards Instance-Level Parser Selection for Cross-Lingual Transfer of Dependency Parsers. Proceedings of COLING.

PDF Poster

(2019). Evaluating resource-lean cross-lingual embedding models in unsupervised retrieval. Proceedings of SIGIR.

PDF Poster

(2019). How to (Properly) Evaluate Cross-Lingual Word Embeddings: On Strong Baselines, Comparative Analyses, and Some Misconceptions. Proceedings of ACL.

PDF Poster

(2018). Unsupervised Cross-Lingual Information Retrieval Using Monolingual Data Only. Proceedings of SIGIR.

PDF Poster

Teaching

Teaching assistant: