E Pavlick - Annual Review of Linguistics, 2022 - annualreviews.org
Deep learning has recently come to dominate computational linguistics, leading to claims of human-level performance in a range of language processing tasks. Like much previous …
The power of human language and thought arises from systematic compositionality—the algebraic ability to understand and produce novel combinations from known components …
Large language models (LLMs) show impressive abilities via few-shot prompting. Commercialized APIs such as OpenAI GPT-3 further increase their use in real-world …
AI is undergoing a paradigm shift with the rise of models (eg, BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We …
We introduce Dynabench, an open-source platform for dynamic dataset creation and model benchmarking. Dynabench runs in a web browser and supports human-and-model-in-the …
Humans can reason compositionally when presented with new tasks. Previous research shows that appropriate prompting techniques enable large language models (LLMs) to …
Despite the recent success of multi-task learning and transfer learning for natural language processing (NLP), few works have systematically studied the effect of scaling up the number …
PW Koh, S Sagawa, H Marklund… - International …, 2021 - proceedings.mlr.press
Distribution shifts—where the training distribution differs from the test distribution—can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild …
The ability to generalize well is one of the primary desiderata for models of natural language processing (NLP), but what 'good generalization'entails and how it should be evaluated is …