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

On the Limits of Learning to Actively Learn Semantic Representations

Omri KoshorekGabriel StanovskyYichu ZhouVivek Srikumar and Jonathan Berant
2019
CoNLL

One of the goals of natural language understanding is to develop models that map sentences into meaning representations. However, training such models requires expensive annotation of complex… 

Y'all should read this! Identifying Plurality in Second-Person Personal Pronouns in English Texts

Gabriel StanovskyRonen Tamari
2019
EMNLP • W-NUT

Distinguishing between singular and plural "you" in English is a challenging task which has potential for downstream applications, such as machine translation or coreference resolution. While formal… 

Universal Adversarial Triggers for Attacking and Analyzing NLP

Eric WallaceShi FengNikhil KandpalSameer Singh
2019
EMNLP

dversarial examples highlight model vulnerabilities and are useful for evaluation and interpretation. We define universal adversarial triggers: input-agnostic sequences of tokens that trigger a… 

Do NLP Models Know Numbers? Probing Numeracy in Embeddings

Eric WallaceYizhong WangSujian LiMatt Gardner
2019
EMNLP

The ability to understand and work with numbers (numeracy) is critical for many complex reasoning tasks. Currently, most NLP models treat numbers in text in the same way as other tokens---they embed… 

AllenNLP Interpret: A Framework for Explaining Predictions of NLP Models

Eric WallaceJens TuylsJunlin WangSameer Singh
2019
EMNLP

Neural NLP models are increasingly accurate but are imperfect and opaque---they break in counterintuitive ways and leave end users puzzled at their behavior. Model interpretation methods ameliorate… 

Balanced Datasets Are Not Enough: Estimating and Mitigating Gender Bias in Deep Image Representations

Tianlu WangJieyu ZhaoMark YatskarVicente Ordonez
2019
ICCV

In this work, we present a framework to measure and mitigate intrinsic biases with respect to protected variables --such as gender-- in visual recognition tasks. We show that trained models… 

Compositional Questions Do Not Necessitate Multi-hop Reasoning

Sewon MinEric WallaceSameer SinghLuke Zettlemoyer
2019
ACL

Multi-hop reading comprehension (RC) questions are challenging because they require reading and reasoning over multiple paragraphs. We argue that it can be difficult to construct large multi-hop RC… 

COMET: Commonsense Transformers for Automatic Knowledge Graph Construction

Antoine BosselutHannah RashkinMaarten SapYejin Choi
2019
ACL

We present the first comprehensive study on automatic knowledge base construction for two prevalent commonsense knowledge graphs: ATOMIC (Sap et al., 2019) and ConceptNet (Speer et al., 2017).… 

HellaSwag: Can a Machine Really Finish Your Sentence?

Rowan ZellersAri HoltzmanYonatan BiskYejin Choi
2019
ACL

Recent work by Zellers et al. (2018) introduced a new task of commonsense natural language inference: given an event description such as "A woman sits at a piano," a machine must select the most… 

The Risk of Racial Bias in Hate Speech Detection

Maarten SapDallas CardSaadia GabrielNoah A. Smith
2019
ACL

We investigate how annotators’ insensitivity to differences in dialect can lead to racial bias in automatic hate speech detection models, potentially amplifying harm against minority populations. We…