conftrace_

character perturbation

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

differentiable tokenization (1) robust language model (1) neural spelling (1) neural network (1) word perturbation (1) text processing (1) model confidence (1) machine comprehension (1) context representation (1) spelling correction (1)

Papers

MANTa: Efficient Gradient-Based Tokenization for End-to-End Robust Language Modeling EMNLP 2022 Evaluating Neural Model Robustness for Machine Comprehension EACL 2021 NeuSpell: A Neural Spelling Correction Toolkit EMNLP 2020