CALO

We work on multimodal generation for task-oriented dialog systems. This work is funded by DARPA. Publications related to this work include:
  • A. Stent and H. Guo. A new data-driven approach for multimedia presentation planning. In Proceedings of EuroIMSA 2005, 2005.
  • A. Stent, M. Marge and M. Singhai. Evaluating evaluation methods for generation in the presence of variation. In Proceedings of CICLing 2005, 2005.
  • H. Zhong and A. Stent. Building surface realizers automatically from corpora using general-purpose tools. In Proceedings of the Workshop on Using Corpora for Natural Language Generation (UCNLG), 2005.

Adaptation in Spoken Dialog

Together with colleagues in the Psychology and Linguistics departments at Stony Brook, we are exploring how humans adapt in dialog, both to other humans and to computers partners. We are exploring adaptations in dialect, lexical choice and syntactic choice. This work is funded by the NSF. See our project web site here. Publications related to this work include:
  • S. Gupta and A. Stent. Automatic evaluation of referring expression generation using corpora. In Proceedings of the Workshop on Using Corpora for Natural Language Generation (UCNLG), 2005.

HearSay

With the Applied Logic Lab at SBU, we are working on HearSay, an audio browser for people with visual disabilities. Publications related to this project include:
  • I. Ramakrishnan, A. Stent and G. Yang. HearSay: Enabling audio browsing on hypertext content. In Proceedings of WWW 2004, 2004.

Machine Translation

With DI Systems, we built a prototype interactive machine translation system for education. This work was funded by DI Systems under the SPIR program.