Allen Institute for AI


Viewing 1-10 of 18 papers
  • Obtaining Faithful Interpretations from Compositional Neural Networks

    Sanjay Subramanian, Ben Bogin, Nitish Gupta, Tomer Wolfson, Sameer Singh, Jonathan Berant, Matt Gardner ACL2020Neural module networks (NMNs) are a popular approach for modeling compositionality: they achieve high accuracy when applied to problems in language and vision, while reflecting the compositional structure of the problem in the network architecture. However, prior work implicitly assumed that the… more
  • QuASE: Question-Answer Driven Sentence Encoding.

    Hangfeng He, Qiang Ning, Dan RothACL2020Question-answering (QA) data often encodes essential information in many facets. This paper studies a natural question: Can we get supervision from QA data for other tasks (typically, non-QA ones)? For example, {\em can we use QAMR (Michael et al., 2017) to improve named entity recognition?} We… more
  • Latent Compositional Representations Improve Systematic Generalization in Grounded Question Answering

    Ben Bogin, Sanjay Subramanian, Matt Gardner, Jonathan BerantarXiv2020Answering questions that involve multi-step reasoning requires decomposing them and using the answers of intermediate steps to reach the final answer. However, state-ofthe-art models in grounded question answering often do not explicitly perform decomposition, leading to difficulties in… more
  • Analyzing Compositionality in Visual Question Answering

    Sanjay Subramanian, Sameer Singh, Matt GardnerNeurIPS • ViGIL Workshop2019Since the release of the original Visual Question Answering (VQA) dataset, several newer datasets for visual reasoning have been introduced, often with the express intent of requiring systems to perform compositional reasoning. Recently, transformer models pretrained on large amounts of images and… more
  • Evaluating Question Answering Evaluation

    Anthony Chen, Gabriel Stanovsky, Sameer Singh, Matt GardnerEMNLP • MRQA Workshop2019As the complexity of question answering (QA) datasets evolve, moving away from restricted formats like span extraction and multiple-choice (MC) to free-form answer generation, it is imperative to understand how well current metrics perform in evaluating QA. This is especially important as existing… more
  • On Making Reading Comprehension More Comprehensive

    Matt Gardner, Jonathan Berant, Hannaneh Hajishirzi, Alon Talmor, Sewon MinEMNLP • MRQA Workshop2019Machine reading comprehension, the task of evaluating a machine’s ability to comprehend a passage of text, has seen a surge in popularity in recent years. There are many datasets that are targeted at reading comprehension, and many systems that perform as well as humans on some of these datasets… more
  • ORB: An Open Reading Benchmark for Comprehensive Evaluation of Machine Reading Comprehension

    Dheeru Dua, Ananth Gottumukkala, Alon Talmor, Sameer Singh, Matt GardnerEMNLP • MRQA Workshop2019Reading comprehension is one of the crucial tasks for furthering research in natural language understanding. A lot of diverse reading comprehension datasets have recently been introduced to study various phenomena in natural language, ranging from simple paraphrase matching and entity typing to… more
  • AllenNLP Interpret: A Framework for Explaining Predictions of NLP Models

    Eric Wallace, Jens Tuyls, Junlin Wang, Sanjay Subramanian, Matthew Gardner, Sameer SinghEMNLP2019
    Best Demo Paper Award
    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 this opacity by providing explanations for specific model predictions. Unfortunately, existing… more
  • Do NLP Models Know Numbers? Probing Numeracy in Embeddings

    Eric Wallace, Yizhong Wang, Sujian Li, Sameer Singh, Matt GardnerEMNLP2019The 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 them as distributed vectors. Is this enough to capture numeracy? We begin by investigating the… more
  • Universal Adversarial Triggers for Attacking and Analyzing NLP

    Eric Wallace, Shi Feng, Nikhil Kandpal, Matthew Gardner, Sameer Singh EMNLP2019dversarial examples highlight model vulnerabilities and are useful for evaluation and interpretation. We define universal adversarial triggers: input-agnostic sequences of tokens that trigger a model to produce a specific prediction when concatenated to any input from a dataset. We propose a… more