Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
文献紹介: Decomposable Neural Paraphrase Generation
Search
Yumeto Inaoka
July 23, 2019
Research
0
190
文献紹介: Decomposable Neural Paraphrase Generation
2019/07/23の文献紹介で発表
Yumeto Inaoka
July 23, 2019
Tweet
Share
More Decks by Yumeto Inaoka
See All by Yumeto Inaoka
文献紹介: Quantity doesn’t buy quality syntax with neural language models
yumeto
1
130
文献紹介: Open Domain Web Keyphrase Extraction Beyond Language Modeling
yumeto
0
180
文献紹介: Self-Supervised_Neural_Machine_Translation
yumeto
0
120
文献紹介: Comparing and Developing Tools to Measure the Readability of Domain-Specific Texts
yumeto
0
130
文献紹介: PAWS: Paraphrase Adversaries from Word Scrambling
yumeto
0
100
文献紹介: Beyond BLEU: Training Neural Machine Translation with Semantic Similarity
yumeto
0
220
文献紹介: EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing
yumeto
0
280
文献紹介: Analyzing the Limitations of Cross-lingual Word Embedding Mappings
yumeto
0
180
文献紹介: Similarity-Based Reconstruction Loss for Meaning Representation
yumeto
1
180
Other Decks in Research
See All in Research
第 2 部 11 章「大規模言語モデルの研究開発から実運用に向けて」に向けて / MLOps Book Chapter 11
upura
0
420
[依頼講演] 適応的実験計画法に基づく効率的無線システム設計
k_sato
0
170
打率7割を実現する、プロダクトディスカバリーの7つの極意(pmconf2024)
geshi0820
0
120
Weekly AI Agents News! 8月号 論文のアーカイブ
masatoto
1
220
Practical The One Person Framework
asonas
1
1.8k
Weekly AI Agents News! 10月号 プロダクト/ニュースのアーカイブ
masatoto
1
150
LLM時代にLabは何をすべきか聞いて回った1年間
hargon24
1
530
文書画像のデータ化における VLM活用 / Use of VLM in document image data conversion
sansan_randd
2
310
研究の進め方 ランダムネスとの付き合い方について
joisino
PRO
56
20k
文献紹介:A Multidimensional Framework for Evaluating Lexical Semantic Change with Social Science Applications
a1da4
1
230
VisFocus: Prompt-Guided Vision Encoders for OCR-Free Dense Document Understanding
sansan_randd
1
340
メタヒューリスティクスに基づく汎用線形整数計画ソルバーの開発
snowberryfield
3
610
Featured
See All Featured
Git: the NoSQL Database
bkeepers
PRO
427
64k
Dealing with People You Can't Stand - Big Design 2015
cassininazir
365
25k
BBQ
matthewcrist
85
9.4k
Designing on Purpose - Digital PM Summit 2013
jponch
116
7k
ReactJS: Keep Simple. Everything can be a component!
pedronauck
665
120k
Large-scale JavaScript Application Architecture
addyosmani
510
110k
Intergalactic Javascript Robots from Outer Space
tanoku
270
27k
[RailsConf 2023 Opening Keynote] The Magic of Rails
eileencodes
28
9.1k
Site-Speed That Sticks
csswizardry
1
180
Save Time (by Creating Custom Rails Generators)
garrettdimon
PRO
28
900
CSS Pre-Processors: Stylus, Less & Sass
bermonpainter
356
29k
Building a Scalable Design System with Sketch
lauravandoore
460
33k
Transcript
Decomposable Neural Paraphrase Generation
https://arxiv.org/abs/1906.09741
• • • •
• • •
•
• • •
• • •
• •
• • = [1 , … , ] • =
[1 , … , ]
• • • ℎ = BiLSTM( ; ℎ−1 , ℎ+1
) • = LSTM ℎ , −1 ; −1 • = GumbelSoftmax( , )
• • = − encoderz (, ) • 1:−1 ,
= − encoderz , 1:−1
• • 1:−1 , = σ 1:−1 , ( |1:−1
, )
• 0 , 1 • = LSTM 0 ; 1
; −1 ; −1 • 1:−1 , = GumbelSoftmax ,
• • ∗ = 0 ∗ = 1
• • ℒ = σ=1 log 1:−1 , + σ=1
log ∗ + σ=1 log ( ∗ 1:−1 ,
• • •
• •
•
• •
• •
• •
• •
• •
• • From 1(best) to 4(worst)
• • • • •