• Publications
  • Influence
Open Information Extraction: The Second Generation
TLDR
The second generation of Open IE systems are described, which rely on a novel model of how relations and their arguments are expressed in English sentences to double precision/recall compared with previous systems such as TEXTRUNNER and WOE. Expand
Towards Coherent Multi-Document Summarization
TLDR
G-FLOW is evaluated on Mechanical Turk, and it is found that it generates dramatically better summaries than an extractive summarizer based on a pipeline of state-of-the-art sentence selection and reordering components, underscoring the value of the joint model. Expand
An analysis of open information extraction based on semantic role labeling
TLDR
This work investigates the use of semantic role labeling techniques for the task of Open IE and compares SRL-based open extractors with TextRunner, an open extractor which uses shallow syntactic analysis but is able to analyze many more sentences in a fixed amount of time and thus exploit corpus-level statistics. Expand
Semantic Role Labeling for Open Information Extraction
TLDR
This work investigates the use of semantic features (semantic roles) for the task of Open IE and finds that SRL-IE is robust to noisy heterogeneous Web data and outperforms TextRunner on extraction quality. Expand
Hierarchical Summarization: Scaling Up Multi-Document Summarization
TLDR
SUMMA produces a hierarchy of relatively short summaries, in which the top level provides a general overview and users can navigate the hierarchy to drill down for more details on topics of interest. Expand
Machine Reading at the University of Washington
TLDR
A unifying approach for machine reading is proposed by bootstrapping from the easiest extractable knowledge and conquering the long tail via a self-supervised learning process that is made scalable by leveraging hierarchical structures and coarse-to-fine inference. Expand
Teaching Classification Boundaries to Humans
TLDR
From their experiments on 54 human subjects learning and performing a pair of synthetic classification tasks via their teaching system, it is found that the authors can achieve the greatest gains with a combination of shaping and the coverage model. Expand
Instance-Driven Attachment of Semantic Annotations over Conceptual Hierarchies
TLDR
This paper introduces a method for converting flat sets of instance-level annotations into hierarchically organized, concept- level annotations, which capture not only the broad semantics of the desired arguments, but also the correct level of generality. Expand
A Rose is a Roos is a Ruusu: Querying Translations for Web Image Search
TLDR
An image search engine, Idiom, is built, which improves the quality of returned images by focusing search on the desired sense, and searches for multiple, automatically chosen translations of the sense in several languages. Expand
Towards Large Scale Summarization
TLDR
This work formalizes the characteristics necessary for good hierarchical summary and provides algorithms to generate them and performs user studies which demonstrate the value of hierarchical summaries over competing methods on datasets much larger than used for traditional summarization. Expand
...
1
2
...