• Publications
  • Influence
The Proposition Bank: An Annotated Corpus of Semantic Roles
TLDR
An automatic system for semantic role tagging trained on the corpus is described and the effect on its performance of various types of information is discussed, including a comparison of full syntactic parsing with a flat representation and the contribution of the empty trace categories of the treebank.
Automatic Labeling of Semantic Roles
TLDR
This work presents a system for identifying the semantic relationships, or semantic roles, filled by constituents of a sentence within a semantic frame, derived from parse trees and hand-annotated training data.
Automatic Labeling of Semantic Roles
TLDR
A system for identifying the semantic relationships, or semantic roles, filled by constituents of a sentence within a semantic frame, based on statistical classifiers trained on roughly 50,000 sentences that were hand-annotated with semantic roles by the FrameNet semantic labeling project.
Effects of disfluencies, predictability, and utterance position on word form variation in English conversation.
TLDR
This study investigates which factors affect the forms of function words, especially whether they have a fuller pronunciation or a more reduced or lenited pronunciation, based on over 8000 occurrences of the ten most frequent English function words in a 4-h sample from conversations from the Switchboard corpus.
A Graph-to-Sequence Model for AMR-to-Text Generation
TLDR
This work introduces a neural graph-to-sequence model, using a novel LSTM structure for directly encoding graph-level semantics, and shows superior results to existing methods in the literature.
Topic-based language models using EM
TLDR
A novel statistical language model to capture topic-related long-range dependencies and an EM algorithm to perform a topic factor decomposition based on a segmented training corpus is proposed.
Corpus Variation and Parser Performance
TLDR
This work examines how other types of text might a ect parser performance, and how portable parsing models are across corpora by comparing results for the Brown and WSJ corpora, and considers which parts of the parser's probability model are particularly tuned to the corpus on which it was trained.
The Necessity of Parsing for Predicate Argument Recognition
TLDR
The effect of parser accuracy on these systems' performance is quantified, and the question of whether a flatter "chunked" representation of the input can be as effective for the purposes of semantic role identification is examined.
Syntactic Features for Evaluation of Machine Translation
TLDR
The results show that adding syntactic information to the evaluation metric improves both sentence-level and corpus-level correlation with human judgments.
N-ary Relation Extraction using Graph-State LSTM
TLDR
This work proposes a graph-state LSTM model, which uses a parallel state to model each word, recurrently enriching state values via message passing, and speeds up computation by allowing more parallelization.
...
...