Articles

We cannot use non-local features with cur- rent major methods of sequence labeling such as CRFs due to concerns about com- plexity. We propose a new perceptron algo- rithm that can use non-local features. Our algorithm allows the use of all types of non-local features whose values are deter- mined …...

This paper proposes a method using the ex- isting Rule-based Machine Translation (RBMT) system as a black box to produce synthetic bilingual corpus, which will be used as training data for the Statistical Ma- chine Translation (SMT) system. We use the existing RBMT system to translate the monolingual corpus into …...

We present a domain-independent unsuper- vised topic segmentation approach based on hybrid document indexing. Lexical chains have been successfully employed to evalu- ate lexical cohesion of text segments and to predict topic boundaries. Our approach is based in the notion of semantic cohesion. It uses spectral embedding to estimate seman- …...

Given multiple translations of the same source sentence, how to combine them to produce a translation that is better than any single system output? We propose a hier- archical system combination framework for machine translation. This framework inte- grates multiple MT systems’ output at the word-, phrase- and sentence- levels. …...

We present a simple history-based model for sentence generation from LFG f-structures, which improves on the accuracy of previous models by breaking down PCFG indepen- dence assumptions so that more f-structure conditioning context is used in the predic- tion of grammar rule expansions. In addi- tion, we present work on …...

To date, work on Non-Local Dependencies (NLDs) has focused almost exclusively on English and it is an open research question how well these approaches migrate to other languages. This paper surveys non-local de- pendency constructions in Chinese as repre- sented in the Penn Chinese Treebank (CTB) and provides an approach …...

This paper describes a probabilistic model for coordination disambiguation integrated into syntactic and case structure analy- sis. Our model probabilistically assesses the parallelism of a candidate coordinate structure using syntactic/semantic similari- ties and cooccurrence statistics. We inte- grate these probabilities into the framework of fully-lexicalized parsing based on large- scale …...

In this paper we propose an instance based method for lexical entailment and apply it to automatic ontology population from text. The approach is fully unsupervised and based on kernel methods. We demonstrate the effectiveness of our technique largely surpassing both the random and most fre- quent baselines and outperforming …...

This paper investigates why the HMMs es- timated by Expectation-Maximization (EM) produce such poor results as Part-of-Speech (POS) taggers. We find that the HMMs es- timated by EM generally assign a roughly equal number of word tokens to each hid- den state, while the empirical distribution of tokens to POS …...

We describe a discriminatively trained se- quence alignment model based on the av- eraged perceptron. In common with other approaches to sequence modeling using per- ceptrons, and in contrast with comparable generative models, this model permits and transparently exploits arbitrary features of input strings. The simplicity of perceptron training lends …...

The Penn Treebank does not annotate within base noun phrases (NPs), commit- ting only to flat structures that ignore the complexity of English NPs. This means that tools trained on Treebank data cannot learn the correct internal structure of NPs. This paper details the process of adding gold-standard bracketing within …...

A key question facing the parsing commu- nity is how to compare parsers which use different grammar formalisms and produce different output. Evaluating a parser on the same resource used to create it can lead to non-comparable accuracy scores and an over-optimistic view of parser performance. In this paper we …...

This paper presents a language-independent probabilistic answer ranking framework for question answering. The framework esti- mates the probability of an individual an- swer candidate given the degree of answer relevance and the amount of supporting evi- dence provided in the set of answer candi- dates for the question. Our approach …...

We study the impact of syntactic and shallow semantic information in automatic classifi- cation of questions and answers and answer re-ranking. We define (a) new tree struc- tures based on shallow semantics encoded in Predicate Argument Structures (PASs) and (b) new kernel functions to exploit the representational power of such …...

The idea of “nugget pyramids” has re- cently been introduced as a refinement to the nugget-based methodology used to evaluate answers to complex questions in the TREC QA tracks. This paper examines data from the 2006 evaluation, the first large-scale de- ployment of the nugget pyramids scheme. We show that …...

In this paper, we propose guided learning, a new learning framework for bidirectional sequence classification. The tasks of learn- ing the order of inference and training the local classifier are dynamically incorporated into a single Perceptron like learning algo- rithm. We apply this novel learning algo- rithm to POS tagging. …...

We describe a new loss function, due to Jeon and Lin (2006), for estimating structured log-linear models on arbitrary features. The loss function can be seen as a (generative) al- ternative to maximum likelihood estimation with an interesting information-theoretic in- terpretation, and it is statistically consis- tent. It is substantially …...

Unsupervised learning of linguistic structure is a difficult problem. A common approach is to define a generative model and max- imize the probability of the hidden struc- ture given the observed data. Typically, this is done using maximum-likelihood es- timation (MLE) of the model parameters. We show using part-of-speech tagging …...

We consider the problem of predictive infer- ence for probabilistic binary sequence label- ing models under F-score as utility. For a simple class of models, we show that the number of hypotheses whose expected F- score needs to be evaluated is linear in the sequence length and present a framework …...

Current phrase-based SMT systems perform poorly when using small training sets. This is a consequence of unreliable translation es- timates and low coverage over source and target phrases. This paper presents a method which alleviates this problem by exploit- ing multiple translations of the same source phrase. Central to our …...