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
サブセット探索を用いた高速なkNNニューラル機械翻訳
Search
Hiroyuki Deguchi
March 22, 2024
Research
0
160
サブセット探索を用いた高速なkNNニューラル機械翻訳
第8回AAMTセミナー
AAMT若手翻訳研究会
最優秀賞
Hiroyuki Deguchi
March 22, 2024
Tweet
Share
More Decks by Hiroyuki Deguchi
See All by Hiroyuki Deguchi
20250226 NLP colloquium: "SoftMatcha: 10億単語規模コーパス検索のための柔らかくも高速なパターンマッチャー"
de9uch1
0
650
20240820: Minimum Bayes Risk Decoding for High-Quality Text Generation Beyond High-Probability Text
de9uch1
0
310
20240226_AAMT-Japio
de9uch1
0
170
Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM’s Translation Capability
de9uch1
0
150
Paper Reading: Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation
de9uch1
0
200
My Research Environmental Setup
de9uch1
0
320
Nearest Neighbor Machine Translation
de9uch1
0
270
Paper Reading - Dynamic Programming Encoding for Subword Segmentation in Neural Machine Translation
de9uch1
0
290
paper reading - Tree Transformer
de9uch1
0
260
Other Decks in Research
See All in Research
生成AI による論文執筆サポート・ワークショップ 論文執筆・推敲編 / Generative AI-Assisted Paper Writing Support Workshop: Drafting and Revision Edition
ks91
PRO
0
120
【SIGGRAPH Asia 2025】Lo-Fi Photograph with Lo-Fi Communication
toremolo72
0
110
製造業主導型経済からサービス経済化における中間層形成メカニズムのパラダイムシフト
yamotty
0
480
学習型データ構造:機械学習を内包する新しいデータ構造の設計と解析
matsui_528
6
3.1k
"主観で終わらせない"定性データ活用 ― プロダクトディスカバリーを加速させるインサイトマネジメント / Utilizing qualitative data that "doesn't end with subjectivity" - Insight management that accelerates product discovery
kaminashi
15
20k
Attaques quantiques sur Bitcoin : comment se protéger ?
rlifchitz
0
140
Sat2City:3D City Generation from A Single Satellite Image with Cascaded Latent Diffusion
satai
4
660
2026.01ウェビナー資料
elith
0
210
大規模言語モデルにおけるData-Centric AIと合成データの活用 / Data-Centric AI and Synthetic Data in Large Language Models
tsurubee
1
490
世界の人気アプリ100個を分析して見えたペイウォール設計の心得
akihiro_kokubo
PRO
66
37k
空間音響処理における物理法則に基づく機械学習
skoyamalab
0
190
Upgrading Multi-Agent Pathfinding for the Real World
kei18
0
200
Featured
See All Featured
WENDY [Excerpt]
tessaabrams
9
36k
brightonSEO & MeasureFest 2025 - Christian Goodrich - Winning strategies for Black Friday CRO & PPC
cargoodrich
3
100
Technical Leadership for Architectural Decision Making
baasie
1
240
Odyssey Design
rkendrick25
PRO
1
500
Making the Leap to Tech Lead
cromwellryan
135
9.7k
Collaborative Software Design: How to facilitate domain modelling decisions
baasie
0
140
Discover your Explorer Soul
emna__ayadi
2
1.1k
How to Ace a Technical Interview
jacobian
281
24k
The Cost Of JavaScript in 2023
addyosmani
55
9.5k
From Legacy to Launchpad: Building Startup-Ready Communities
dugsong
0
140
How Software Deployment tools have changed in the past 20 years
geshan
0
32k
YesSQL, Process and Tooling at Scale
rocio
174
15k
Transcript
𝒌
◼ ⚫ ⚫ ◼ ⚫ (Zhang+, NAACL2018; Gu+, AAAI2018; Khandelwal+,
ICLR2021) ▶ (Nagao, 1984) ▶ ⚫ 𝑘 (Khandelwal+, ICLR2021) ▶ ▶ ▶ Guiding Neural Machine Translation with Retrieved Translation Pieces (Zhang+, NAACL2018) Search Engine Guided Neural Machine Translation (Gu+, AAAI2018) Nearest Neighbor Machine Translation (Khandelwal+, ICLR2021) A framework for a mechanical translation between Japanese and English by analogy principle (Nagao, 1984)
◼ ◼ ⚫ ⚫
𝒌 (Khandelwal+, ICLR2021) ◼ ⚫ ⚫ ⚫ ◼ ⚫ ▶
⚫ ▶ ≈ Nearest Neighbor Machine Translation (Khandelwal+, ICLR2021) 𝒙 𝒚
𝒌 (Khandelwal+, ICLR2021) 𝒌𝑖 ∈ ℝ𝐷 𝑓 𝒙, 𝒚<𝑡 ∈
ℝ𝐷 Nearest Neighbor Machine Translation (Khandelwal+, ICLR2021) ◼ 𝑘 ◼ ⚫ ⚫ 𝑝𝑘NN 𝑦𝑡 𝒙, 𝒚<𝑡 ∝ 𝑖=1 𝑘 𝟙𝑦𝑡=𝑣𝑖 exp − 𝒌𝑖 − 𝑓 𝒙, 𝒚<𝑡 2 2 𝜏 ◼ 𝑘
𝒌 ◼ (Martins+, EMNLP2022) ◼ (Meng+, ACLFindings2022) ⚫ 𝑘 𝑘
𝜆 = 0.5 𝑘 = 16 Chunk-based Nearest Neighbor Machine Translation (Martins+, EMNLP2022) Fast Nearest Neighbor Machine Translation (Meng+, ACL Findings2022)
𝒌 ◼ 𝑘 ◼ ⚫ 𝑘 (Matsui+, ACMMM2018) ⚫ 𝑘
𝑘 𝑘 Reconfigurable Inverted Index (Matsui+, ACMMM2018) 𝒌
◼ ⚫ 𝑘 ⚫ 𝑘 ◼ ◼ 𝑘
𝑛 𝑘 1 1 1 1 1 1 1 1
1
𝑛 𝑘 1 1 1 1 1 1 1 1
1
𝑛 𝑘 1 1 1 1 1 1 1 1
1
⚫ ⚫ ⚫ ⚫ ⚫ 𝑘 𝜆 = 0.5 𝑘
= 16 𝑛 = 56
𝑘 𝑘 ◼ 𝑘 ⚫ ▶ ⚫ ▶
◼ 𝑘 𝒌 𝒌
◼ ⚫ 𝑘
𝒌 𝒌 ◼ ⚫ ⚫ ◼ 𝑘 ⚫ ⚫ ◼
⚫
⚫ ⚫ ▶ ⚫ ▶