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

MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers

Krishna PillutlaSwabha SwayamdiptaRowan ZellersZ. Harchaoui
2021
NeurIPS

As major progress is made in open-ended text generation, measuring how close machine-generated text is to human language remains a critical open problem. We introduce MAUVE , a comparison measure… 

MERLOT: Multimodal Neural Script Knowledge Models

Rowan ZellersXiming LuJack HesselYejin Choi
2021
NeurIPS

As humans, we understand events in the visual world contextually, performing multimodal reasoning across time to make inferences about the past, present, and future. We introduce MERLOT, a model… 

Natural Adversarial Objects

Felix LauNishant SubramaniSasha HarrisonRosanne Liu
2021
NeurIPS 2021 Data Centric AI Workshop

Although state-of-the-art object detection methods have shown compelling performance, models often are not robust to adversarial attacks and out-of-distribution data. We introduce a new dataset,… 

NaturalProofs: Mathematical Theorem Proving in Natural Language

S. WelleckJiachen LiuRonan Le BrasKyunghyun Cho
2021
NeurIPS

Understanding and creating mathematics using natural mathematical language – the mixture of symbolic and natural language used by humans – is a challenging and important problem for driving progress… 

One Question Answering Model for Many Languages with Cross-lingual Dense Passage Retrieval

Akari AsaiXinyan YuJungo KasaiHanna Hajishirzi
2021
NeurIPS

We present CORA, a Cross-lingual Open-Retrieval Answer Generation model that can answer questions across many languages even when language-specific annotated data or knowledge sources are… 

Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing

Sarah Wiegreffe and Ana Marasović
2021
NeurIPS Datasets & Benchmarks

Explainable NLP (ExNLP) has increasingly focused on collecting human-annotated explanations. These explanations are used downstream in three ways: as data augmentation to improve performance on a… 

Bridging the Imitation Gap by Adaptive Insubordination

Luca WeihsUnnat JainJordi SalvadorA. Schwing
2021
arXiv

Why do agents often obtain better reinforcement learning policies when imitating a worse expert? We show that privileged information used by the expert is marginalized in the learned agent policy,… 

Iconary: A Pictionary-Based Game for Testing Multimodal Communication with Drawings and Text

Christopher ClarkJordi SalvadorDustin SchwenkAli Farhadi
2021
arXiv

Communicating with humans is challenging for AIs because it requires a shared understanding of the world, complex semantics (e.g., metaphors or analogies), and at times multimodal gestures (e.g.,… 

Specializing Multilingual Language Models: An Empirical Study

Ethan C. ChauNoah A. Smith
2021
EMNLP • Workshop on Multilingual Representation Learning

Pretrained multilingual language models have become a common tool in transferring NLP capabilities to low-resource languages, often with adaptations. In this work, we study the performance,… 

Towards Personalized Descriptions of Scientific Concepts

Sonia K. MurthyDaniel KingTom HopeDoug Downey
2021
EMNLP 2021 • WiNLP

A single scientific concept can be described in many different ways, and the most informative description depends on the audience. In this paper, we propose generating personalized scientific…