
Masahiro Kaneko
fairness
grammatical error correction
social bias
evaluation
word embeddings
interpretability
language model
machine learning
masked language model
instruction
efficiency
knn
reranking
meta-embedding
sense embeddings
12
presentations
12
number of views
SHORT BIO
I am a postdoctoral researcher at Tokyo Institute of Technology.
Presentations

Reducing Sequence Length by Predicting Edit Spans with Large Language Models
Masahiro Kaneko and 1 other author

The Impact of Debiasing on the Performance of Language Models in Downstream Tasks is Underestimated
Masahiro Kaneko and 2 other authors

Exploring Effectiveness of GPT-3 in Grammatical Error Correction: A Study on Performance and Controllability in Prompt-Based Methods
Mengsay Loem and 3 other authors

Comparing Intrinsic Gender Bias Evaluation Measures without using Human Annotated Examples
Masahiro Kaneko and 2 other authors

NAIST-NICT-TIT WMT22 General MT Task Submission
Hiroyuki Deguchi and 7 other authors

IMPARA: Impact-Based Metric for GEC Using Parallel Data
Koki Maeda and 2 other authors

Debiasing isn't enough! -- On the Effectiveness of Debiasing MLMs and their Social Biases in Downstream Tasks
Masahiro Kaneko and 2 other authors

ExtraPhrase: Efficient Data Augmentation for Abstractive Summarization
Mengsay Loem and 3 other authors

Sense Embeddings are also Biased -- Evaluating Social Biases in Static and Contextualised Sense Embeddings
Yi Zhou and 2 other authors

Interpretability for Language Learners Using Example-Based Grammatical Error Correction
Masahiro Kaneko and 3 other authors

Unmasking the Mask – Evaluating Social Biases in Masked Language Models
Masahiro Kaneko and 1 other author

Gender Bias in Meta-Embeddings
Masahiro Kaneko and 2 other authors