Skip to main content ->
Ai2

Research - Papers

Explore a selection of our published work on a variety of key research challenges in AI.

Filter papers

BERTese: Learning to Speak to BERT

Adi HavivJonathan BerantA. Globerson
2021
EACL

Large pre-trained language models have been shown to encode large amounts of world and commonsense knowledge in their parameters, leading to substantial interest in methods for extracting that… 

Bootstrapping Relation Extractors using Syntactic Search by Examples

Matan EyalAsaf AmramiHillel Taub-TabibYoav Goldberg
2021
EACL

The advent of neural-networks in NLP brought with it substantial improvements in supervised relation extraction. However, obtaining a sufficient quantity of training data remains a key challenge. In… 

Evaluating the Evaluation of Diversity in Natural Language Generation

Guy TevetJonathan Berant
2021
EACL

Despite growing interest in natural language generation (NLG) models that produce diverse outputs, there is currently no principled method for evaluating the diversity of an NLG system. In this… 

First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT

Benjamin MullerYanai ElazarBenoît SagotDjamé Seddah
2021
EACL

Multilingual pretrained language models have demonstrated remarkable zero-shot crosslingual transfer capabilities. Such transfer emerges by fine-tuning on a task of interest in one language and… 

Formalizing Trust in Artificial Intelligence: Prerequisites, Causes and Goals of Human Trust in AI

Alon JacoviAna MarasovićTim MillerYoav Goldberg
2021
FAccT

Trust is a central component of the interaction between people and AI, in that 'incorrect' levels of trust may cause misuse, abuse or disuse of the technology. But what, precisely, is the nature of… 

Leap-Of-Thought: Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge

Alon TalmorOyvind TafjordPeter ClarkJonathan Berant
2020
NeurIPS • Spotlight Presentation

To what extent can a neural network systematically reason over symbolic facts? Evidence suggests that large pre-trained language models (LMs) acquire some reasoning capacity, but this ability is… 

It's not Greek to mBERT: Inducing Word-Level Translations from Multilingual BERT

Hila GonenShauli RavfogelYanai ElazarYoav Goldberg
2020
EMNLP • BlackboxNLP Workshop

Recent works have demonstrated that multilingual BERT (mBERT) learns rich cross-lingual representations, that allow for transfer across languages. We study the word-level translation information… 

Unsupervised Distillation of Syntactic Information from Contextualized Word Representations

Shauli RavfogelYanai ElazarJacob GoldbergerYoav Goldberg
2020
EMNLP • BlackboxNLP Workshop

Contextualized word representations, such as ELMo and BERT, were shown to perform well on various semantic and syntactic task. In this work, we tackle the task of unsupervised disentanglement… 

The Extraordinary Failure of Complement Coercion Crowdsourcing

Yanai ElazarVictoria BasmovShauli RavfogelReut Tsarfaty
2020
EMNLP • Insights from Negative Results in NLP Workshop

Crowdsourcing has eased and scaled up the collection of linguistic annotation in recent years. In this work, we follow known methodologies of collecting labeled data for the complement coercion… 

A Novel Challenge Set for Hebrew Morphological Disambiguation and Diacritics Restoration

Avi ShmidmanJoshua GuedaliaShaltiel ShmidmanReut Tsarfaty
2020
Findings of EMNLP

One of the primary tasks of morphological parsers is the disambiguation of homographs. Particularly difficult are cases of unbalanced ambiguity, where one of the possible analyses is far more…