Clippers is our forum for informal discussion of all issues related to computational linguistics: from work in progress of visitors and people in the department, over presentation of new papers, to practical concerns such as hints on the use of CL related software tools.
Everyone with an interest in computational linguistics is most welcome!
To see what happened in previous quarters of Clippers, you can check out the pages of some previous quarters: Fall 05 Spring 05 Winter 05, Autumn 04, Spring 04, Winter 04, Autumn 03, Spring 03, Winter 03, Autumn 02, Spring 02, Autumn 01
When and where: Fridays at 130- 248 in 340 Central Classrooms.
Important: Please be sure to subscribe to our local computational linguistics mailing list on which all Clippers sessions and talks are announced.
The plan, as usual, is to start each session with 5-10 minutes on whatever someone wants to bring up and then to continue with the following topics:
|
Date |
Presenter |
Topic |
|---|---|---|
|
6 Jan |
Xiaofei Lu |
Practice Job Talk |
|
13 January |
Donna Byron, Chris Brew, maybe Tianfeng |
Professional Day Chris's slides |
|
20 January |
Ilana Bromberg |
Arabic Information Retrieval: Past, Present, Future |
|
27 January |
Eric Fosler-Lussier and friends (at short notice, thankyou!) |
We'll discuss (mostly informally) some of the current goings on in the Automatic Speech Attribute Transcription (ASAT) project, including an explanation of why Eric keeps muttering to himself "Dirichlet, Dirichlet, Dirichlet..." |
|
3 February |
Kirk Baker |
English words in Japanese |
|
10 February |
cancelled |
|
|
17 February |
Markus Dickinson |
The Errors of our ways |
|
24 February |
Crystal Nakatsu, Mike White |
Learning to Say It Well: Reranking Realizations by Predicted Synthesis Quality |
|
3 March |
Donna Byron |
This talk describes results on a content planning task for a spoken dialog agent that provides incremental, real-time navigation instructions. The task consists of a direction-giver steering his partner, the follower, to a series of target referents within a large interior space. The content planner must determine whether the follower is in a spatial context from which a description of the referent is likely to be understood. This choice is driven by spatial context features derived from the follower's position, such as his view angle and distance from the target and the number of similar items in the current field of view. Trained on a set of 2-party human dialogs, our algorithm achieved 85% precision in matching the performance of human subjects. |
|
10 March |
|
|