In the current formulation of TESLA-CELAB, two n-grams X and Y are either synonyms which completely match each other, or are completely unrelated.

Compared to BLEU, TESLA allows more sophisticated weighting of n-grams and measures of word similarity including synonym relations.

The covered n-gram matching rule is then able to award tricky n-grams such as TE, Ti, /1\ [E], 1/13 [IE5 and i9}.

For example, between ¥_l—?fi_$ and ¥_5l?, higher-order n-grams such as and still have no match, and will be penalized accordingly, even though ¥_l—?fi_5lk and ¥_5l?

N-grams such as which cross natural word boundaries and are meaningless by themselves can be particularly tricky.

Two n-grams are connected if they are identical, or if they are identified as synonyms by Cilin.

Notice that all n-grams are put in the same matching problem regardless of n, unlike in translation evaluation metrics designed for European languages.

This enables us to designate n-grams with different values of n as synonyms, such as (n = 2) and 5!k (n = 1).

We represent each utterance a as a vector wu of Nu word n-grams (segments), wuj, each of which are chosen from a vocabulary W of fixed-size V. We use entity lists obtained from web sources (explained next) to identify segments in the corpus.

Web n-Grams (G).

Our vocabulary consists of n—grams and segments (phrases) in utterances that are extracted using web n-grams and entity lists of §3.

* Web n-Gram Context Base Measure (2%): As explained in §3, we use the web n-grams as additional information for calculating the base measures of the Dirichlet topic distributions.

In (1) we assume that entities (E) are more indicative of the domain compared to other n-grams (G) and should be more dominant in sampling decision for domain topics.

During Gibbs sampling, we keep track of the frequency of draws of domain, dialog act and slot indicating n-grams wj, in M D, M A and MS matrices, respectively.

As the source of Web information, we use the Google n-grams corpus (Brants and Franz, 2006) which contains English n-grams (n = 1 to 5) and their Web frequency counts, derived from nearly 1 trillion word tokens and 95 billion sentences.

Using the n-grams corpus (for n = l to 5), we collect co-occurrence Web-counts by allowing a varying number of wildcards between hl and hg in the query.

2These clusters are derived form the V2 Google n-grams corpus.

translation hypothesis to compute the numbers of the reference n-grams .

Both BLEU and PORT perform matching of n-grams up to n = 4.

In all tuning experiments, both BLEU and PORT performed lower case matching of n-grams up to n = 4.

The BLEU-tuned and Qmean-tuned systems generate similar numbers of matching n-grams, but Qmean-tuned systems produce fewer n-grams (thus, shorter translations).

This is mainly due to the additional minimum support constraint we added which discards many noisy lexical entries from infrequently seen n-grams .

and the corresponding navigation plan, we first segment the instruction into word tokens and construct n-grams from them.

From the corresponding navigation plan, we find all connected subgraphs of size less than or equal to m. We then update the co-occurrence counts between all the n-grams w and all the connected subgraphs 9.

We also update the counts of how many examples we have encountered so far and counts of the n-grams w and subgraphs 9.

Table 3: MWE identification with CRF: base are the features corresponding to token properties and word n-grams .

Word n-grams .

POS n-grams .

Field bigrams/trigrams Analogously to the lexical features mentioned above, we introduce a series of nonlocal features that capture field n-grams , given a specific record.

Local and nonlocal information (e.g., word n-grams , long-

The l-BEST system has some grammaticality issues, which we avoid by defining features over lexical n-grams and repeated words.

In this paper, we only tried Dice coefficient of n-grams and symmetrical sentence level BLEU as similarity measures.

Tl(e,e') is the propagating probability in equation (8), with the similarity measure Sim(e,e') defined as the Dice coefficient over the set of all n-grams in e and those in e'.

where N Grn(x) is the set of n-grams in string x, and Dice (A, B) is the Dice coefficient over sets A and B:

While traditional LMs use word n-grams , where the n — 1 previous words predict the next word, newer models integrate long-span information in making decisions.

For example, incorporating long-distance dependencies and syntactic structure can help the LM better predict words by complementing the predictive power of n-grams (Chelba and Jelinek, 2000; Collins et al., 2005; Filimonov and Harper, 2009; Kuo et al., 2009).

Structured language modeling incorporates syntactic parse trees to identify the head words in a hypothesis for modeling dependencies beyond n-grams .

Feature templates such as rule n-grams and rule shapes only work if iterative mixing (algorithm 3) or feature selection (algorithm 4) are used.

Such features include rule ids, rule-local n-grams , or types of rule shapes.

Rule n-grams: These features identify n-grams of consecutive items in a rule.