Smoothing and Shrinking the Sparse Seq2Seq Search Space

Smoothing and Shrinking the Sparse Seq2Seq Search Space

18 March 2021

André F. T. Martins

ArXiv (abs)PDF HTML

Papers citing "Smoothing and Shrinking the Sparse Seq2Seq Search Space"

7 / 7 papers shown

Title
Multi-Objective Hyperparameter Selection via Hypothesis Testing on Reliability Graphs Amirmohammad Farzaneh Osvaldo Simeone 516 0 0 22 Jan 2025
A Continuum of Generation Tasks for Investigating Length Bias and Degenerate Repetition Darcey Riley David Chiang 71 5 0 19 Oct 2022
Efficient Methods for Natural Language Processing: A Survey Marcos Vinícius Treviso Ji-Ung Lee Tianchu Ji Betty van Aken Qingqing Cao ... Emma Strubell Niranjan Balasubramanian Leon Derczynski Iryna Gurevych Roy Schwartz 156 114 0 31 Aug 2022
The Implicit Length Bias of Label Smoothing on Beam Search Decoding Bowen Liang Pidong Wang Yuan Cao 67 1 0 02 May 2022
Characterizing and addressing the issue of oversmoothing in neural autoregressive sequence modeling Ilia Kulikov M. Eremeev Kyunghyun Cho 66 8 0 16 Dec 2021
Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation Bryan Eikema Wilker Aziz 89 48 0 10 Aug 2021
What Do You Get When You Cross Beam Search with Nucleus Sampling? Uri Shaham Omer Levy 83 10 0 20 Jul 2021