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
76
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
20240820: Minimum Bayes Risk Decoding for High-Quality Text Generation Beyond High-Probability Text
de9uch1
0
170
20240226_AAMT-Japio
de9uch1
0
97
Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM’s Translation Capability
de9uch1
0
110
Paper Reading: Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation
de9uch1
0
130
My Research Environmental Setup
de9uch1
0
240
Nearest Neighbor Machine Translation
de9uch1
0
200
Paper Reading - Dynamic Programming Encoding for Subword Segmentation in Neural Machine Translation
de9uch1
0
250
paper reading - Tree Transformer
de9uch1
0
200
Dependency-Based Self-Attention for Transformer NMT (RANLP2019)
de9uch1
0
63
Other Decks in Research
See All in Research
Neural Fieldの紹介
nnchiba
1
570
ECCV2024読み会: Minimalist Vision with Freeform Pixels
hsmtta
1
360
EBPMにおける生成AI活用について
daimoriwaki
0
260
クロスセクター効果研究会 熊本都市交通リノベーション~「車1割削減、渋滞半減、公共交通2倍」の実現へ~
trafficbrain
0
350
情報処理学会関西支部2024年度定期講演会「自然言語処理と大規模言語モデルの基礎」
ksudoh
10
2.4k
博士学位論文予備審査 / Scaling Telemetry Workloads in Cloud Applications: Techniques for Instrumentation, Storage, and Mining
yuukit
1
1.6k
論文紹介: COSMO: A Large-Scale E-commerce Common Sense Knowledge Generation and Serving System at Amazon (SIGMOD 2024)
ynakano
1
320
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
220
精度を無視しない推薦多様化の評価指標
kuri8ive
1
350
Segment Any Change
satai
2
100
文化が形作る音楽推薦の消費と、その逆
kuri8ive
0
220
CoRL2024サーベイ
rpc
1
1.4k
Featured
See All Featured
Documentation Writing (for coders)
carmenintech
67
4.6k
The Success of Rails: Ensuring Growth for the Next 100 Years
eileencodes
44
7k
The Myth of the Modular Monolith - Day 2 Keynote - Rails World 2024
eileencodes
20
2.4k
A designer walks into a library…
pauljervisheath
205
24k
Principles of Awesome APIs and How to Build Them.
keavy
126
17k
What’s in a name? Adding method to the madness
productmarketing
PRO
22
3.2k
[Rails World 2023 - Day 1 Closing Keynote] - The Magic of Rails
eileencodes
33
2k
Fireside Chat
paigeccino
34
3.2k
StorybookのUI Testing Handbookを読んだ
zakiyama
28
5.4k
Improving Core Web Vitals using Speculation Rules API
sergeychernyshev
5
200
VelocityConf: Rendering Performance Case Studies
addyosmani
327
24k
Building Flexible Design Systems
yeseniaperezcruz
328
38k
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 ⚫ ⚫ ◼
⚫
⚫ ⚫ ▶ ⚫ ▶