ISCOL 2017

Computational Linguistics and Natural Language Processing are active research and development fields in Israel today, both in academia and industry. ISCOL is a venue for exchanging ideas, reporting on work in progress and established results, forming collaborations, and advancing connections between academia and industry. ISCOL is also a friendly stage for students for their first appearance in this community.

Program

09:00-09:05: Opening Words

09:05-10:25: Morning Session (Session Chair: Michael Elhadad)

Compositional Semantic Parsing of Instructions in Unseen Domains
Ofer Givoli and Roi Reichart
Neural Semantic Parsing over Multiple Knowledge-bases
Jonathan Herzig and Jonathan Berant
A Consolidated Open Knowledge Representation for Multiple Texts
Rachel Wities, Vered Shwartz, Gabriel Stanovsky, Meni Adler, Ori Shapira, Shyam Upadhyay, Dan Roth, Eugenio Martinez Camara, Iryna Gurevych and Ido Dagan
Deep Learning for Preprocessing Historical Hebrew Texts: Error Correction, Vocalization and Abbreviation Expansion
Avi Shmidman, Shaltiel Shmidman, Moshe Koppel and Yoav Goldberg

10:25-11:00: Coffee Break

11:00-12:00: Invited Talk by Inbal Arnon (Session Chair: Omri Abend)

More Than Words: Psycholinguistic and Computational Investigations of the Building Blocks of Language

Abstract:

Why are children better language learners than adults despite being worse at a range of other cognitive tasks? Here, I explore the role of multiword sequences in explaining L1–L2 differences in learning and language use more generally. Words are often seen as the basic building blocks of language learning and processing (e.g., Pinker, 1991), a perception that is reflected in the focus on words as units of processing in both the psycholinguistic and computational literature. However, there is growing theoretical interest in and empirical evidence for the role of multiword units in language. I draw on developmental, psycholinguistic and computational findings to argue that multiword sequences, like words, are integral building blocks for language and play a particularly important role in first language learning. These findings show that children use multiword units in learning and that adult native speakers still rely on them. In contrast, adult L2 learners make less use of such units in learning, a pattern that can explain some of the difference between child and adult language learning. I discuss implications for computational models of language learning and NLP more generally.

Short bio:

Inbal Arnon is an associate professor in the Psychology Department at Hebrew University of Jerusalem. Her main interests are first language acquisition, learning theory, psycholinguistics, and the differences between first and second language learning. She has a BA in linguistics from Tel-Aviv university (Summa cum Laude), an MSc in psycholinguistics from the University of Edinburgh (with distinction), and a PhD in Linguistics from Stanford University. Her research lies at the intersection of Linguistics, Psychology, and Cognitive Science and uses a variety of experimental methods to explore how language is learned and how learning changes as a function of prior knowledge and experience.

12:00-14:00: Poster Session and Light Lunch (see list of posters)

14:00-14:40: An Open Discussion on the Future of the Mila Center: Alon Itai (chair), Shuly Wintner

14:40-16:00: Afternoon Session (Session Chair: Yoav Goldberg)

Too Many Corrections: Semantic Reference-less Evaluation for Grammatical Error Correction
Leshem Choshen and Omri Abend
Integrating Deep Linguistic Features in Factuality Prediction over Unified Datasets
Gabriel Stanovsky, Judith Eckle-Kohler, Yevgeniy Puzikov, Ido Dagan and Iryna Gurevych
ConStance: Modeling Annotation Contexts to Improve Stance Classification
Kenneth Josseph, Lisa Friedland, William Hobbs, David Lazer and Oren Tsur
Found in Translation: Reconstructing Phylogenetic Language Trees from Translations
Ella Rabinovich, Noam Ordan and Shuly Wintner

16:00-16:30: Coffee Break

16:30-17:30: Evening Session (Session Chair: Jonathan Berant)

Controlling Linguistic Style Aspects in Neural Language Generation
Jessica Ficler and Yoav Goldberg
Unsupervised Query-Focused Multi-Document Summarization
Guy Feigenblat, Haggai Roitman, Odellia Boni and David Konopnicki
Query Focused Abstractive Summarization: Adapting Sequence-to-Sequence Models with Query Relevance
Tal Baumel, Matan Eyal and Michael Elhadad

Poster Presentations:

Towards String-to-Tree Neural Machine Translation
Roee Aharoni and Yoav Goldberg
A Transition-Based Directed Acyclic Graph Parser for Universal Conceptual Cognitive Annotation
Daniel Hershcovich, Omri Abend and Ari Rappoport
Interactive Abstractive Summarization for Event News Tweets
Ori Shapira, Hadar Ronen, Adler Meni, Yael Amsterdamer, Judit Bar-Ilan and Ido Dagan
Semantic Helper for Text Compose
Boris Pevzner
Acquiring Predicate Paraphrases from News Tweets
Vered Shwartz, Gabriel Stanovsky and Ido Dagan
Learning Agreement with Deep Neural Networks
Jean-Philippe Bernardy and Shalom Lappin
Mutual Convergence of Mutlimodal Linguistic Properties in Spoken Dialogue Systems
Eran Raveh, Ingmar Steiner, Iona Gessinger and Bernd Möbius
Unsupervised Event Detection in Twitter
Yuval Shachaf, Gennadi Lembersky and Roy Klein
Evaluating Semantic Parsing against a Simple Web-based Question Answering Model
Alon Talmor, Mor Geva and Jonathan Berant
Semantic Structure-Based Evaluation Metrics for Sentence Simplification
Elior Sulem, Omri Abend and Ari Rappoport
Recognizing Mentions of Adverse Drug Reaction in Social Media Using Knowledge-Infused Recurrent Models
Gabriel Stanovsky, Daniel Gruhl and Pablo Mendes
Automatic Detection of Translation Direction
Ilia Sominsky and Shuly Wintner
Thomson Reuters Self-Service Classification
Ehud Azikri and Hila Zarosim
Interactive Semantic Parsing on Semi-Structured Tables
Tomer Wolfson
Floating Agenda Parser
Dor Muhlgay and Jonathan Berant
Reinforcing Abstractive Text Summarization with Reading-Comprehension
Matan Eyal, Tal Baumel and Michael Elhadad
Improving Claim Stance Classification with Lexical Knowledge Expansion and Context Utilization
Roy Bar-Haim, Lilach Edelstein, Charles Jochim and Noam Slonim
Multi-Label Classification on Patient Notes
Tal Baumel, Jumana Nassour-Kassis, Noemie Elhadad and Michael Elhadad
Crafting a Lexicon of Referential Expressions for NLG Applications
Ariel Gutman, Alexandros Andre Chaaraoui and Pascal Fleury
Where's My Head? Task Definition, Dataset and Initial Results for Fused Head Recovery
Yanai Elazar and Yoav Goldberg
Contextualization and Regularization of Word Embeddings for Machine Comprehension
Shimi Salant and Jonathan Berant
Semantically Motivated Hebrew Verb-Noun Multi-Word Expressions Identification
Chaya Liebeskind and Yaakov Hacohen-Kerner
Detection of Native Language Based on Micro Messages
Yohay Nahari, Maayan Matook and Oren Tsur
Causal Factors of Language Change Under Control - the Case of Frequency Bias in Laws of Semantic Change
Haim Dubossarsky, Eitan Grossman and Daphna Weinshall
Rare Word Embeddings with Sub-word Similarity Based Search: Better Representations without More Knowledge, Data, or Training
Clayton Greenberg, Mittul Singh, Youssef Oualil, and Dietrich Klakow

Time and Location

Monday, September 25, 2017

The Brindell and Milton Gottlieb Auditorium (A300),
The Rothberg Family buildings,
Rachel and Selim Benin School of Computer Science and Engineering,
Edmond J. Safra Campus, Jerusalem

Campus map »