Subword & Character Level Models in NLP

The 2nd Workshop on Subword and Character level models in NLP (SCLeM) will be held at NAACL 2018 in New Orleans, Louisiana USA.

The first edition of SCLeM was held at EMNLP 2017.

Following the success of the first edition, with over 100 participants, the workshop provides a forum for discussing recent advances as well as future directions on sub-word and character-level natural language processing and representation learning. A detailed discussion on ideas and motivation behind this workshop are described here.

We thank Microsoft Research for sponsoring our workshop in part.

Invited Speakers

• Jacob Eisenstein, Georgia Tech
• Wang Ling, DeepMind
• Graham Neubig, CMU
• Barbara Plank, IT University of Copenhagen
• Brian Roark, Google

Topics

• tokenization-free models
• character-level machine translation
• character-ngram information retrieval
• transfer learning for character-level models
• models of within-token and cross-token structure
• NL generation (of words not seen in training etc)
• out of vocabulary words
• morphology & segmentation
• relationship b/w morphology & character-level models
• stemming and lemmatization
• inflection generation
• orthographic productivity
• form-meaning representations
• true end-to-end learning
• spelling correction
• efficient and scalable character-level models

Anti-harassment Policy

Our workshop highly values the open exchange of ideas, the freedom of thought and expression, and respectful scientific debate. We support and uphold the ACL Anti-Harassment policy, and any workshop participant should feel free to contact any of the NAACL Board members or Priscilla Rasmussen, in case of any issues.