Litcius/Paper detail

How Context Affects Language Models' Factual Predictions

Fabio Petroni, Patrick Lewis, Aleksandra Piktus, Tim Rocktäschel, Yuxiang Wu, Alexander Miller, Sebastian Riedel

2020UCL Discovery (University College London)41 citations

Abstract

When pre-trained on large unsupervised textual corpora, language models are able to
\nstore and retrieve factual knowledge to some extent, making it possible to use them directly for zero-shot cloze-style question answering. However, storing factual knowledge in a
\nfixed number of weights of a language model clearly has limitations. Previous approaches
\nhave successfully provided access to information outside the model weights using supervised architectures that combine an information retrieval system with a machine reading
\ncomponent. In this paper, we go a step further and integrate information from a retrieval
\nsystem with a pre-trained language model in a purely unsupervised way. We report that
\naugmenting pre-trained language models in this way dramatically improves performance
\nand that the resulting system, despite being unsupervised, is competitive with a supervised machine reading baseline. Furthermore, processing query and context with different
\nsegment tokens allows BERT to utilize its Next Sentence Prediction pre-trained classifier
\nto determine whether the context is relevant or not, substantially improving BERT’s zeroshot cloze-style question-answering performance and making its predictions robust to noisy
\ncontexts.

Topics & Concepts

Computer scienceArtificial intelligenceLanguage modelNatural language processingQuestion answeringSentenceClassifier (UML)Context (archaeology)Reading (process)Machine learningLinguisticsBiologyPhilosophyPaleontologyTopic ModelingNatural Language Processing TechniquesMultimodal Machine Learning Applications