Grusha Prasad

发表

How much harder are hard garden-path sentences than easy ones?

Tal Linzen, Grusha Prasad, Tal Linzen, 2019, CogSci.

Counterfactual Interventions Reveal the Causal Effect of Relative Clause Representations on Agreement Prediction

Yoav Goldberg, Tal Linzen, Shauli Ravfogel, 2021, CONLL.

Using Priming to Uncover the Organization of Syntactic Representations in Neural Language Models

Tal Linzen, Marten van Schijndel, Grusha Prasad, 2019, CoNLL.

Dynabench: Rethinking Benchmarking in NLP pdf

Zhiyi Ma, Christopher Potts, Douwe Kiela, 2021, NAACL.

To what extent do human explanations of model behavior align with actual model behavior? pdf

Adina Williams, Douwe Kiela, Mohit Bansal, 2020, BLACKBOXNLP.

Rapid syntactic adaptation in self-paced reading: Detectable, but only with many participants.

Tal Linzen, Grusha Prasad, 2019, Journal of experimental psychology. Learning, memory, and cognition.

SPR mega-benchmark shows surprisal tracks construction- but not item-level difficulty

Brian, Tal Linzen, Suhas Arehalli, 2022 .

Rapid syntactic adaptation in self-paced reading: Detectable, but only with many participants.

Tal Linzen, Grusha Prasad, 2019, Journal of experimental psychology. Learning, memory, and cognition.

Large-scale benchmark yields no evidence that language model surprisal explains syntactic disambiguation difficulty

Tal Linzen, Suhas Arehalli, Grusha Prasad, 2024, Journal of Memory and Language.