News

Joint Prediction of Word Alignment with Alignment Types
22 Jun 2016

At the June 22nd lab meeting (11:30am), Anahita will talk about Joint Prediction of Word Alignment with Alignment Types.

Abstract: Probabilistic models for word alignment do not distinguish between different types of word alignment links. We provide a new probabilistic model for word alignment in which alignments are associated with linguistically motivated alignment types. We provide a novel task of joint prediction of word alignment and alignment types.

Andrei's GPU Parsing Presentation
19 May 2016

On May 25th 11:30am, Andrei will give a presentation on the CKY algorithm and Canny, Hall, and Klein’s 2013 paper “A Multi-Teraflop Constituency Parser using GPUs.”

NW-NLP 2016
18 May 2016

Several lab members will be traveling to the North West NLP Workshop NW-NLP 2016 in Seattle, USA on Friday May 20th.

The workshop is hosted by Amazon, Inc. at Suite 700, 1220 Howell Street, Seattle WA.

The following papers from the SFU Natlang Lab will be presented there:

  • Talk at 10am: Improving Statistical Machine Translation with a Multilingual Paraphrase Database, Ramtin M. Seraj, Maryam Siahbani, and Anoop Sarkar
  • Posters at 12:30pm:
    • Learning Segmentations that Balance Latency versus Quality in Spoken Language Translation, Hassan S. Shavarani, Maryam Siahbani, Ramtin Mehdizadeh Seraj, Anoop Sarkar.
    • Evaluating classification accuracy of funding for research in the Brazilian Federal Official Gazette, Paulo Marques and Fred Popowich.
    • Recommending Alternate Reading Materials Using Probabilistic Topic Models, Lydia Odilinye and Fred Popowich.
    • Visualizing Political Press Briefings For Changes In Stance Between Agents, Milan Tofiloski and Fred Popowich.

Also presenting from SFU Linguistics:

  • Talk at 3:30pm: Mapping different rhetorical relation annotations: a proposal, Farah Benamara and Maite Taboada.

Jasneet practice talk
18 May 2016

On May 18th 11:30am, Jasneet will give a practice talk for his Masters thesis defence on the topic “BILINGUAL LANGUAGE MODELS USING WORD EMBEDDINGS FOR MACHINE TRANSLATION”.

Jasneet MSc thesis defence
18 May 2016

On May 19th 2pm in TASC1 9204, Jasneet will defend his Masters thesis on the topic “BILINGUAL LANGUAGE MODELS USING WORD EMBEDDINGS FOR MACHINE TRANSLATION”.

Abstract

Bilingual language models (Bi-LMs) refer to language models over pairs of words in source and target languages in a machine translation task. When translating from source to target language, the decoder in phrase-based machine translation system segments the source sentence into phrases and then translates each phrase to the target language. While decoding each phrase, the decoder does not have sufficient information about source words that are outside the phrase under consideration. Bi-LMs have been used to tackle this problem. Bi-LMs are estimated by first creating bi-token sequences using word alignments over a parallel corpus. We propose the use of bilingual word embeddings to deal with the large number of bi-token types in a bi-token language model. Our approach outperforms previous work with an increase of 1.4 BLEU points in our machine translation experiments.

M.Sc. Examining Committee:

  • Dr. Anoop Sarkar, Co-Senior Supervisor
  • Dr. Fred Popowich, Co-Senior Supervisor
  • Dr. Jiannan Wang, Examiner
  • Dr. Ryan Shea, Chair

Recent Publications