R
Regina Barzilay
Researcher at Massachusetts Institute of Technology
Publications - 327
Citations - 26243
Regina Barzilay is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Automatic summarization & Computer science. The author has an hindex of 77, co-authored 297 publications receiving 20544 citations. Previous affiliations of Regina Barzilay include Columbia University & IBM.
Papers
More filters
Using lexical chains for text summarization
Regina Barzilay,Michael Elhadad +1 more
TL;DR: Empirical results on the identification of strong chains and of significant sentences are presented in this paper, and plans to address short-comings are briefly presented.
Journal ArticleDOI
A Deep Learning Approach to Antibiotic Discovery
Jonathan M. Stokes,Kevin Yang,Kyle Swanson,Wengong Jin,Andres Cubillos-Ruiz,Nina M. Donghia,Craig R. MacNair,Shawn French,Lindsey A. Carfrae,Zohar Bloom-Ackermann,Victoria M. Tran,Anush Chiappino-Pepe,Ahmed H. Badran,Ian W. Andrews,Ian W. Andrews,Ian W. Andrews,Emma J. Chory,George M. Church,Eric D. Brown,Tommi S. Jaakkola,Regina Barzilay,James J. Collins +21 more
TL;DR: A deep neural network capable of predicting molecules with antibacterial activity is trained and a molecule from the Drug Repurposing Hub-halicin- is discovered that is structurally divergent from conventional antibiotics and displays bactericidal activity against a wide phylogenetic spectrum of pathogens.
Journal ArticleDOI
Modeling local coherence: An entity-based approach
TL;DR: This article re-conceptualize coherence assessment as a learning task and shows that the proposed entity-grid representation of discourse is well-suited for ranking-based generation and text classification tasks.
Journal ArticleDOI
Analyzing Learned Molecular Representations for Property Prediction.
Kevin Yang,Kyle Swanson,Wengong Jin,Connor W. Coley,Philipp Eiden,Hua Gao,Angel Guzman-Perez,Timothy Hopper,Brian Kelley,Miriam Mathea,Andrew Palmer,Volker Settels,Tommi S. Jaakkola,Klavs F. Jensen,Regina Barzilay +14 more
TL;DR: In this article, a graph convolutional model that consistently matches or outperforms models using fixed molecular descriptors as well as previous graph neural architectures on both public and proprietary data sets is presented.
Proceedings ArticleDOI
Rationalizing Neural Predictions
TL;DR: The authors combine two modular components, generator and encoder, which are trained to operate well together to extract pieces of input text as justifications, tailored to be short and coherent, yet sufficient for making the same prediction.