Our research projects represent a comprehensive approach to decompositional semantics, each tackling a fundamental aspect of meaning representation through innovative annotation frameworks.

9 Projects
200K+ Total Annotations
19 Publications

Decomp Toolkit

Comprehensive Python toolkit for working with Universal Decompositional Semantics datasets and models.

1 paper
Open Source
Explore Project

Decompositional Parsing

Joint syntactic and semantic parsing models that learn to extract graph structures and attributes from text.

3 papers
Open Source
Explore Project

PredPatt

Universal predicate-argument extraction that bridges deep syntax and shallow semantics across languages.

2 papers
Open Source
Explore Project

Semantic Proto-Roles

Fine-grained thematic role properties that capture how events relate to their participants through decomposed semantic features.

2 datasets
5 papers
Explore Project

Factuality

Determining whether events described in text actually happened, incorporating speaker confidence and evidentiality.

2 datasets
3 papers
Explore Project

Genericity

Distinguishing generic statements about kinds from specific episodic events through multi-dimensional annotation.

2 datasets
1 paper
Explore Project

Event Structure

Capturing complex event structures including subparts, participants, and their relationships over time.

3 datasets
1 paper
Explore Project

Time

Real-valued temporal relations and event durations for constructing precise document-level timelines.

1 dataset
2 papers
Explore Project

Word Sense

Word sense disambiguation through simple binary questions about meaning in context.

1 dataset
1 paper
Explore Project