Home > Unit > UMIACS

UMIACS

Cache Transition Systems for Semantic Parsing

Abstract: We describe a transition system that generalizes standard transition-based dependency parsing techniques to generate a graph rather than a tree.  Our system includes a cache with fixed size m, and we characterize the relationship between the parameter m and the class of graphs that can be produced through the graph-theoretic concept of tree decomposition.  We train a sequence-to-sequence neural model based on this system to parse text into Abstract Meaning Representation (AMR).

TitleA family of neural models for voice query understanding on an entertainment platform


Title: Event Semantics in Text Constructions, Vision, and Human-Robot Dialogue

Title: SCRIPTS: a System for Cross Language Information Processing,Translation and Summarization

Abstract: This presentation will give an overview of the research conducted by CLIP students and facuty to develop a System for Cross Language Information Processing, Translation and Summarization, as part of the IARPA MATERIAL program.

CLIP lab members get 5 minutes each to talk about their summer research. All are welcome to attend.

 

Language Science Center faculty -  Philip Resnik (Linguistics, UMIACS), Marine Carpuat (Computer Science, UMIACS), and Hal Daume (Computer Science, UMIACS) - join Douglas Oard (iSchool) on a four-year $14.4M Intelligence Advanced Research Projects Activity (IARPA) grant.

Title: Semantic and Stylistic Variations in Machine Translation

Abstract: While parallel texts represent invaluable resources for machine translation, they inevitably introduce biases in the cross-lingual mappings learned by machine translation models.  In addition to the domain bias and translationese bias studied in past work, we argue that another form of bias arises from subtle choices in content and style made by translators to appropriately convey the meaning of the source to their target audience.

Pages

Subscribe to UMIACS