activity
20172022
most citedSource-side Prediction for Neural Headline Generation

12 citations · 25 across the 7 of their papers we have counts for

collaborators

12 papers

cs.LG2022

Diverse Lottery Tickets Boost Ensemble from a Single Pretrained Model

Sosuke Kobayashi, Shun Kiyono, Jun Suzuki +1

Ensembling is a popular method used to improve performance as a last resort. However, ensembling multiple models finetuned from a single pretrained model has been not very effectiv…

cs.CL2021

SHAPE: Shifted Absolute Position Embedding for Transformers

Shun Kiyono, Sosuke Kobayashi, Jun Suzuki +1

Position representation is crucial for building position-aware representations in Transformers. Existing position representations suffer from a lack of generalization to test data…

cs.CL20214 cited

Rethinking Perturbations in Encoder-Decoders for Fast Training

Sho Takase, Shun Kiyono

We often use perturbations to regularize neural models. For neural encoder-decoders, previous studies applied the scheduled sampling (Bengio et al., 2015) and adversarial perturbat…

cs.CL2021

Pseudo Zero Pronoun Resolution Improves Zero Anaphora Resolution

Ryuto Konno, Shun Kiyono, Yuichiroh Matsubayashi +2

Masked language models (MLMs) have contributed to drastic performance improvements with regard to zero anaphora resolution (ZAR). To further improve this approach, in this study, w…

cs.CL20202 cited

An Empirical Study of Contextual Data Augmentation for Japanese Zero Anaphora Resolution

Ryuto Konno, Yuichiroh Matsubayashi, Shun Kiyono +3

One critical issue of zero anaphora resolution (ZAR) is the scarcity of labeled data. This study explores how effectively this problem can be alleviated by data augmentation. We ad…

cs.CL2020

A Self-Refinement Strategy for Noise Reduction in Grammatical Error Correction

Masato Mita, Shun Kiyono, Masahiro Kaneko +2

Existing approaches for grammatical error correction (GEC) largely rely on supervised learning with manually created GEC datasets. However, there has been little focus on verifying…