LETA, as part of Team RIGA, wins the AMR PARSING 2016 trophy at SemEval 2016

SUMMA at SemEval-2016 competition on Meaning Representation Parsing

SemEval-2016, the International Workshop on Semantic Evaluation, was held June 16-17, 2016 at the NAACL HLT 2016 conference in San Diego, CA. SUMMA consortium partners UCL, University of Sheffield and LETA participated in the SemEval-2016 Task 8 on Meaning Represention Parsing. New to this competition was AMR entity grounding via wikification, which pushed the semantic parsing boundaries toward the multi-document scope. Advancing the state-of-art in Abstract Meaning Representation (AMR) parsing is vital for storyline  summarisation envisioned within the SUMMA project.

Team RIGA (LETA and University of Latvia) achieved the top result in the competition. The inaugural trophy (pictured) was awarded to the two best teams ranked according to the following Smatch-scores:

  1. RIGA (University of Latvia, IMCS; LETA): 0.6196
  2. Brandeis/cx/RPI (Brandeis University; Boulder Learning Inc.; Rensselaer Polytechnic Institute): 0.6195
  3. ICL-HD (Ruprecht-Karls-Universit¨at Heidelberg): 0.6005
  4. UCL+Sheffield (University College London; University of Sheffield): 0.5983
  5. M2L (Kyoto University): 0.5952
  6. CMU (Carnegie Mellon University; University of Washington): 0.5636
  7. CU-NLP (OK Robot Go, Ltd.; University of Colorado): 0.5566
  8. UofR    (University of Rochester): 0.4985
  9. MeaningFactory (University of Groningen): 0.4702
  10. CLIP@UMD (University of Maryland): 0.4370
  11. DynamicPower (National Institute for Japanese Language and Linguistics): 0.3706
    Andreas Vlachos (University of Sheffield) and Guntis Barzdins (LETA) at SemEval-2016

    Andreas Vlachos (University of Sheffield) and Guntis Barzdins (LETA) at SemEval-2016