Dr Andreas Vlachos

MEng General Engineering Liaison

Room number: G33 Regent Court
Telephone: +44 (0) 114 222 1960

Personal website: staffwww.dcs.shef.ac.uk/people/A.Vlachos/

ORCID | Google scholar

Selected publications | All publications

Dr Andreas Vlachos



Dr Andreas Vlachos is a lecturer at the University of Sheffield, working on the intersection of Natural Language Processing and Machine Learning. Previously he was a postdoc at the Machine Reading group at UCL working with Sebastian Riedel and the BBC R&D team. Before that he was a member of the NLIP group at the University of Cambridge working with Stephen Clark and further back a postdoc at the University of Wisconsin-Madison working with Mark Craven. He holds a PhD from the University of Cambridge and an MSc from the University of Edinburgh.

Other professional activities and achievements:

  • 2017 Area chair for the conference on Empirical Methods in Natural Language Processing (EMNLP)
  • 2017 Area chair for the European Chapter of the Association for Computational Linguistics (EACL) conference
  • 2014 – now Grant Reviewer for the Research Foundation - Flanders (FWO)
  • 2014 – now Standing review committee for the Transactions of the Association for Computational Linguistics
  • 2006 – now Member of the Association for Computational Linguistics (ACL)
  • 2015 – now Member of the International Machine Learning Society (IMLS)



Dr Andreas' research focuses on the following areas:

Automated Fact-Checking

Prime minster said: ”Our government has halved youth unemployment!” True or False? Fact checking is one of the main tasks performed by journalists, especially in an era in which information sources abound. The first paper discussed the main challenges, namely the open domain nature of the task and the importance of context: temporal, geographical, conversational. This was followed up by a distantly supervised approach for fact-checking simple claims about statistical properties (with Sebastian Riedel)

Imitation Learning for Structured Prediction

Imitation learning is a paradigm originally developed in robotics that has been applied successfully to a variety of structured prediction tasks in NLP. Intuitively, it decomposes the usually complex output (e.g. a graph) to a sequence of actions that construct it. These actions are predicted by a suitalby trained policy. This framework has the advantage of being able to learn policies with non-decomposable loss functions without explicit enumeration of the output search space and has been applied successfully to a variety of applications, including information extraction (biomedical or web-based) and semantic parsing. (with Isabelle Augenstein and Jason Naradowsky)

Domain-independent Natural Language Generation

Currently weather sites host data (temperature, wind speed, humidity) concerning a multitude of cities and areas around the world. However, only a few locations feature textual weather forecasts as well. To automatically generate forecasts from the data itself, the group aims to develop a domain-independent Natural Language Generation (NLG) framework by imitating generation policies from unaligned corpora. This will be applied to a variety of domains, as well as weather reports and data obtained from the Met Office. (EPSRC-funded with Gerasimos Lampouras, Sebastian Riedel and the Interaction Lab)


Current Grants