Proceedings of the main conference on Human Language Technology Conference of the North American Chapter of the Association of Computational Linguistics -

Abstract

It is with pleasure that we preface the publications of the 2006 Human Language Technology conference --- North American chapter of the Association for Computational Linguistics annual meeting (HLTNAACL 2006). The conference has a number of formats by which refereed work can be presented: full papers, short papers (either as a talk or poster), and demonstrations. As befits this multi-disciplinary conference, papers were submitted across the three topics of computational linguistics, information retrieval and speech recognition. This year, 257 full papers were submitted and 62 accepted (25% acceptance rate), 127 short papers submitted and 52 accepted (41% rate). It is pleasing to report that these numbers mark a strong increase in submissions compared to the last HLT NAACL conference run in 2004.The selection of the high quality submissions in these proceedings was the product of a two tiered reviewing system. The three PC chairs selected 28 senior program committee (PC) members, who are internationally recognized for their subject expertise. This group constituted the top tier of the PC. Each of the members selected a group of reviewers to review both the full and short submitted papers. The complete PC numbered around 250. Three reviewers and one senior PC person were assigned per paper. Reviewing was double blinded. The senior PC oversaw the reviewing process, helped resolve any disputes, and at the end produced, for each paper, an overview of the reviewers' comments along with a preliminary decision on whether the submission should be accepted or not. These decisions formed the basis of discussion at a program committee meeting. Separate PC meetings were held for full and short papers. For full, a one day meeting was held at IBM Research Watson, NY; for short papers, a telephone conference call was held between the three PC chairs.The senior PC also nominated candidates for best paper and best student paper, the two selected for the prizes were chosen by the PC chairs working in conjunction with the senior PCs. The papers that won were "Probabilistic Context-Free Grammar Induction Based on Structural Zeros" by Mehryar Mohri and Brian Roark and "Prototype-Driven Learning for Sequence Models" by Aria Haghighi and Dan Klein. Congratulations to them both.