(This is the introduction to the proceedings)

This volume contains the papers accepted for presentation at the 2001 Conference on Empirical Methods in Natural Language Processing (EMNLP 2001), a SIGDAT meeting held at Carnegie Mellon University on June 3 and 4, 2001, just preceding the Second Meeting of the North American Chapter of the ACL.

Along with papers on traditional SIGDAT areas and related fields, we solicited papers addressing this year's conference theme, ``What Works and What Doesn't: Successes and Challenges''. We were looking for both papers showing the success of empirical methods in and across application settings, and papers exposing some of the challenges that still lie before the community by carefully identifying and exploring current limitations. We are very pleased with the breadth and depth of the responses to this call that this year's papers represent.

To further reflection on this year's theme, the conference program also includes an invited talk on the subject, plus two panels: one debating the efficacy of the Expectation-Maximization algorithm (``When does EM work?''), and one on industrial perspectives on what works and what doesn't in natural language technology. Thanks very much to our invited speaker, Eric Brill, and panelists, Adam Berger, Eugene Charniak, David Evans, Joshua Goodman, Lynette Hirschman, Kevin Knight, Ted Pedersen, and Stefan Riezler; their participation contributes greatly to what we think is an exciting and interesting program.

Organizing a meeting like this is a bit of an effort -- indeed, colleagues, in observing our labors, dubbed the enterprise ``the U.S.S. EMNLP''. Luckily, we had an excellent crew on board, with forty program committee members and additional reviewers hailing from fifteen different countries around the world. From an initial slate of 65 submissions, they selected the papers in this proceedings (representing 8 countries in Asia, Europe, and North America) in a double-blind refereeing process. We deeply and sincerely thank all who reviewed: we would have run aground without their efforts.

Additionally, a number of others pitched in at various crucial times to help keep things on course. David Yarowsky lent us his considerable expertise to handle the proceedings printing process. Rie Kubota Ando and Rebecca Hwa graciously donated their time during certain deadline-induced periods of stress. Tammy Howe and Cindy Robinson provided invaluable administrative support. Claire Cardie, Pascale Fung, Pierre Isabelle, and Hinrich Schuetze, all of whom have been at the helm previously, gave us much-appreciated advice and perspective, and Giorgio Satta was kind enough to make available Perl scripts for processing the reviews. We would also like to thank Alon Lavie for handling the local arrangements, Jodi Hauck for being a wonderful liaison to Omnipress, and Priscilla Rasmussen, the ACL Business Manager, through whom all things are possible.

We most gratefully acknowledge the support of our sponsor, the Intelligent Information Systems Institute, a joint Cornell University/Air Force Research Laboratory organization devoted to research in compute- and data-intensive methods for intelligent decision making systems. Carla P. Gomes and Marcy E. Rosenkrantz, Director and Associate Director, provided a very generous amount of vital funding for us, and we have the utmost appreciation for this gift.

Finally, we thank Kenneth Church and David Yarowsky, who convinced us to embark on this enterprise in the first place. They have both been extremely supportive, and served as inexhaustible sources of wise advice. They deserve the full twenty-one gun salute!

Lillian Lee
Donna Harman
May 2001

Back to EMNLP 2001 home page