A more precise analysis of punctuation for broad-coverage surface realization with CCG

Proceedings of the 11th International Conference on Natural Language Generation

2018

Self Cite

Hypertagging, or supertagging for surface realization, is the process of assigning lexical categories to nodes in an input semantic graph. Previous work has shown that hypertagging significantly increases realization speed and quality by reducing the search space of the realizer. Building on recent work using LSTMs to improve accuracy on supertagging for parsing, we develop an LSTM hypertagging method for OpenCCG, an open source NLP toolkit for CCG. Our results show significant improvements in both hypertagging accuracy and downstream realization performance.

Section: Surface Realization With Openccgmentioning

confidence: 99%

LSTM Hypertagging

Fu¹,

Proceedings of the 11th International Conference on Natural Language Generation

2018

Self Cite

“…Our starting point is an enhanced version of the CCGbank (Hockenmaier and Steedman, 2007)-a corpus of CCG derivations derived from the Penn Treebank-with Propbank (Palmer et al, 2005) roles projected onto it (Boxwell and White, 2008). To engineer a grammar from this corpus suitable for realization with OpenCCG, the derivations are first revised to reflect the lexicalized treatment of coordination and punctuation assumed by the multi-modal version of CCG that is implemented in OpenCCG (White and Rajkumar, 2008). Further changes are necessary to support semantic dependencies rather than surface syntactic ones; in particular, the features and unification constraints in the categories related to semantically empty function words such complementizers, infinitivalto, expletive subjects, and case-marking prepositions are adjusted to reflect their purely syntactic status.…”

Section: Realization From An Enhanced Ccgbankmentioning

confidence: 99%

“…Note that the baseline for Section 23 uses 4-grams and a filter for balanced punctuation(White and Rajkumar, 2008), unlike the other reported configurations, which would explain the somewhat smaller increase seen with this section.6 Scripts for running these tests are available at http://projectile.sv.cmu.edu/research/ public/tools/bootStrap/tutorial.htm…”

mentioning

confidence: 99%

Perceptron reranking for CCG realization

Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing Volume 1 - EMNLP '09

Rajkumar

2009

Self Cite

This paper shows that discriminative reranking with an averaged perceptron model yields substantial improvements in realization quality with CCG. The paper confirms the utility of including language model log probabilities as features in the model, which prior work on discriminative training with log linear models for HPSG realization had called into question. The perceptron model allows the combination of multiple n-gram models to be optimized and then augmented with both syntactic features and discriminative n-gram features. The full model yields a stateof-the-art BLEU score of 0.8506 on Section 23 of the CCGbank, to our knowledge the best score reported to date using a reversible, corpus-engineered grammar.

“…In the figure, nodes correspond to discourse referents labeled with lexical predicates, and dependency relations between nodes encode argument structure (gold standard CCG lexical categories are also shown); note that semantically empty function words such as infinitival-to are missing. The grammar is extracted from a version of the CCGbank (Hockenmaier and Steedman, 2007) enhanced for realization; the enhancements include: better analyses of punctuation (White and Rajkumar, 2008); less error prone handling of named entities Figure 1: Example OpenCCG semantic dependency input for he has a point he wants to make, with gold standard lexical categories for each node and assignment of consistent semantic roles across diathesis alternations (Boxwell and White, 2008), using PropBank (Palmer et al, 2005).…”

Section: Introductionmentioning

confidence: 99%

That's Not What I Meant! Using Parsers to Avoid Structural Ambiguities in Generated Text

Duan

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

2014

Self Cite

We investigate whether parsers can be used for self-monitoring in surface realization in order to avoid egregious errors involving "vicious" ambiguities, namely those where the intended interpretation fails to be considerably more likely than alternative ones. Using parse accuracy in a simple reranking strategy for selfmonitoring, we find that with a stateof-the-art averaged perceptron realization ranking model, BLEU scores cannot be improved with any of the well-known Treebank parsers we tested, since these parsers too often make errors that human readers would be unlikely to make. However, by using an SVM ranker to combine the realizer's model score together with features from multiple parsers, including ones designed to make the ranker more robust to parsing mistakes, we show that significant increases in BLEU scores can be achieved. Moreover, via a targeted manual analysis, we demonstrate that the SVM reranker frequently manages to avoid vicious ambiguities, while its ranking errors tend to affect fluency much more often than adequacy.