We present a dynamic programming algorithm for efficient constrained inference in semantic role labeling. The algorithm tractably captures a majority of the structural constraints exam-ined by prior work in this area, which has re-sorted to either approximate methods or off-the-shelf integer linear programming solvers. In ad-dition, it allows training a globally-normalized log-linear model with respect to constrained conditional likelihood. We show that the dy-namic program is several times faster than an off-the-shelf integer linear programming solver, while reaching the same solution. Furthermore, we show that our structured model results in significant improvements over its local counter-part, achieving state-of-the-art results on both PropBank- and FrameNet-annotated corpora. 1

We present a new method for semantic role labeling in which arguments and seman-tic roles are jointly embedded in a shared vector space for a given predicate. These embeddings belong to a neural network, whose output represents the potential func-tions of a graphical model designed for the SRL task. We consider both local and structured learning methods and ob-tain strong results on standard PropBank and FrameNet corpora with a straightfor-ward product-of-experts model. We fur-ther show how the model can learn jointly from PropBank and FrameNet annotations to obtain additional improvements on the smaller FrameNet dataset. 1

Abstract It is in PropBank's ARGM annotation of clausal adjuncts that sentential semantics meets discourse relation annotation in the Penn Discourse TreeBank. This paper discusses complementarities between the two annotation systems: How PropBank ARGM annotation can be used to seed annotation of additional discourse relations in the PDTB, and how PDTB annotation can be used to refine or enrich PropBank ARGM annotation.

Modelling linguistic phenomena requires highly structured and complex data representations. Document representation frameworks (DRFs) provide an interface to store and retrieve multiple annotation layers over a document. Researchers face a difficult choice: using a heavy-weight DRF or implement a custom DRF. The cost is substantial, either learning a new complex system, or continually adding features to a home-grown system that risks overrunning its original scope. We introduce DOCREP, a lightweight and efficient DRF, and compare it against existing DRFs. We discuss our design goals and implementations in C++, Python, and Java. We transform the OntoNotes 5 corpus using DOCREP and UIMA, providing a quantitative comparison, as well as discussing modelling trade-offs. We conclude with qualitative feedback from researchers who have used DOCREP for their own projects. Ultimately, we hope DOCREP is useful for the busy researcher who wants the benefits of a DRF, but has better things to do than to write one. 1