Can Transformer be Too Compositional? Analysing Idiom Processing in
Neural Machine Translation

Can Transformer be Too Compositional? Analysing Idiom Processing in Neural Machine Translation

30 May 2022

Verna Dankers

Christopher G. Lucas

Papers citing "Can Transformer be Too Compositional? Analysing Idiom Processing in Neural Machine Translation"

12 / 12 papers shown

Title
Geometric Signatures of Compositionality Across a Language Model's Lifetime Jin Hwa Lee Thomas Jiralerspong Lei Yu Yoshua Bengio Emily Cheng CoGe 84 0 0 02 Oct 2024
Compositional Generalization with Grounded Language Models Sondre Wold Étienne Simon Lucas Georges Gabriel Charpentier Egor V. Kostylev Erik Velldal Lilja Øvrelid KELM 31 0 0 07 Jun 2024
Divergences between Language Models and Human Brains Yuchen Zhou Emmy Liu Graham Neubig Michael J. Tarr Leila Wehbe 32 1 0 15 Nov 2023
Does the "most sinfully decadent cake ever" taste good? Answering Yes/No Questions from Figurative Contexts Geetanjali Rakshit Jeffrey Flanigan ELM 11 1 0 24 Sep 2023
LEACE: Perfect linear concept erasure in closed form Nora Belrose David Schneider-Joseph Shauli Ravfogel Ryan Cotterell Edward Raff Stella Biderman KELM MU 41 102 0 06 Jun 2023
Finding Memo: Extractive Memorization in Constrained Sequence Generation Tasks Vikas Raunak Arul Menezes 35 13 0 24 Oct 2022
Understanding Transformer Memorization Recall Through Idioms Adi Haviv Ido Cohen Jacob Gidron R. Schuster Yoav Goldberg Mor Geva 28 48 0 07 Oct 2022
Are Representations Built from the Ground Up? An Empirical Examination of Local Composition in Language Models Emmy Liu Graham Neubig CoGe 15 10 0 07 Oct 2022
The paradox of the compositionality of natural language: a neural machine translation case study Verna Dankers Elia Bruni Dieuwke Hupkes CoGe 162 75 0 12 Aug 2021
On Compositional Generalization of Neural Machine Translation Yafu Li Yongjing Yin Yulong Chen Yue Zhang 153 44 0 31 May 2021
The Bottom-up Evolution of Representations in the Transformer: A Study with Machine Translation and Language Modeling Objectives Elena Voita Rico Sennrich Ivan Titov 193 181 0 03 Sep 2019
What you can cram into a single vector: Probing sentence embeddings for linguistic properties Alexis Conneau Germán Kruszewski Guillaume Lample Loïc Barrault Marco Baroni 201 882 0 03 May 2018