Papers

Learn more about AI2's Lasting Impact Award
AI2 Israel
All Years
Viewing 1-10 of 89 papers
  • CommonsenseQA 2.0: Exposing the Limits of AI through Gamification

    Alon Talmor, Ori Yoran, Ronan Le Bras, Chandrasekhar Bhagavatula, Yoav Goldberg, Yejin Choi, Jonathan Berant NeurIPS2021 Constructing benchmarks that test the abilities of modern natural language un1 derstanding models is difficult – pre-trained language models exploit artifacts in 2 benchmarks to achieve human parity, but still fail on adversarial examples and make 3 errors… more
  • Back to Square One: Bias Detection, Training and Commonsense Disentanglement in the Winograd Schema

    Yanai Elazar, Hongming Zhang, Yoav Goldberg, Dan RothEMNLP2021 The Winograd Schema (WS) has been proposed as a test for measuring commonsense capabilities of models. Recently, pre-trained language model-based approaches have boosted performance on some WS benchmarks but the source of improvement is still not clear. We… more
  • Contrastive Explanations for Model Interpretability

    Alon Jacovi, Swabha Swayamdipta, Shauli Ravfogel, Yanai Elazar, Yejin Choi, Yoav GoldbergEMNLP2021 Contrastive explanations clarify why an event occurred in contrast to another. They are more inherently intuitive to humans to both produce and comprehend. We propose a methodology to produce contrastive explanations for classification models by modifying the… more
  • Parameter Norm Growth During Training of Transformers

    William Merrill, Vivek Ramanujan, Yoav Goldberg, Roy Schwartz, Noah A. Smith EMNLP2021 The capacity of neural networks like the widely adopted transformer is known to be very high. Evidence is emerging that they learn successfully due to inductive bias in the training routine, typically some variant of gradient descent (GD). To better… more
  • Transformer Feed-Forward Layers Are Key-Value Memories

    Mor Geva, R. Schuster, Jonathan Berant, Omer LevyEMNLP2021 Feed-forward layers constitute two-thirds of a transformer model’s parameters, yet their role in the network remains underexplored. We show that feed-forward layers in transformer-based language models operate as key-value memories, where each key correlates… more
  • What's in your Head? Emergent Behaviour in Multi-Task Transformer Models

    Mor Geva, Uri Katz, Aviv Ben-Arie, Jonathan BerantEMNLP2021 The primary paradigm for multi-task training in natural language processing is to represent the input with a shared pre-trained language model, and add a small, thin network (head) per task. Given an input, a target head is the head that is selected for… more
  • Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies

    Mor Geva, Daniel Khashabi, Elad Segal, Tushar Khot, Dan Roth, Jonathan BerantTACL2021 A key limitation in current datasets for multi-hop reasoning is that the required steps for answering the question are mentioned in it explicitly. In this work, we introduce STRATEGYQA, a question answering (QA) benchmark where the required reasoning steps… more
  • Few-Shot Question Answering by Pretraining Span Selection

    Ori Ram, Yuval Kirstain, Jonathan Berant, A. Globerson, Omer LevyACL2021 In a number of question answering (QA) benchmarks, pretrained models have reached human parity through fine-tuning on an order of 100,000 annotated questions and answers. We explore the more realistic few-shot setting, where only a few hundred training… more
  • Neural Extractive Search

    Shaul Ravfogel, Hillel Taub-Tabib, Yoav GoldbergACL • Demo Track2021 Domain experts often need to extract structured information from large corpora. We advocate for a search paradigm called “extractive search”, in which a search query is enriched with capture-slots, to allow for such rapid extraction. Such an extractive search… more
  • Question Decomposition with Dependency Graphs

    Matan Hasson, Jonathan BerantACL-IJCNLP • 15th Workshop on Structured Prediction for NLP2021 QDMR is a meaning representation for complex questions, which decomposes questions into a sequence of atomic steps. While stateof-the-art QDMR parsers use the common sequence-to-sequence (seq2seq) approach, a QDMR structure fundamentally describes labeled… more
AI2 Israel
All Years