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:Improving Machine Translation Quality Esti...
Search
Taichi Aida
July 03, 2019
Technology
88
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
文献紹介:Improving Machine Translation Quality Estimation with Neural Network Features
Taichi Aida
July 03, 2019
More Decks by Taichi Aida
See All by Taichi Aida
意味を表すベクトル表現を用いたテキスト分析
a1da4
0
140
スウェーデン滞在報告
a1da4
0
31
PhD Defence: Considering Temporal and Contextual Information for Lexical Semantic Change Detection
a1da4
1
300
文献紹介:A Multidimensional Framework for Evaluating Lexical Semantic Change with Social Science Applications
a1da4
1
400
YANS2024:目指せ国際会議!「ネットワーキングの極意(国際会議編)」
a1da4
0
330
言語処理学会30周年記念事業留学支援交流会@YANS2024:「学生のための短期留学」
a1da4
1
450
新入生向けチュートリアル:文献のサーベイv2
a1da4
18
12k
文献紹介:Isotropic Representation Can Improve Zero-Shot Cross-Lingual Transfer on Multilingual Language Models
a1da4
0
240
文献紹介:WhitenedCSE: Whitening-based Contrastive Learning of Sentence Embeddings
a1da4
1
390
Other Decks in Technology
See All in Technology
Terraformモジュールは、なぜ「魔境」化するのか
hayama17
2
220
noUncheckedIndexedAccess、3時間、1万円。 / noUncheckedIndexedAccess, 3 Hours, 10,000 JPY.
kaonavi
1
340
Platform engineering for developers, architects & the rest of us (AI agents)
danielbryantuk
0
190
AIソロプレナー時代に2ヶ月で20人増員した事業創造会社の開発組織の話
miyatakoji
0
140
あなたの AI ワークスペースに、 専門コーダーを連れてくる - Amazon Quick Desktop 最新情報
kawaji_scratch
1
110
生成 AI × MCP で切り拓く次世代 SRE!自律型運用への挑戦と開発者体験の進化
_awache
0
170
エンジニアリング戦略の作り方 / Crafting Engineering Strategy
iwashi86
12
3.8k
Mastering Ruby Box
tagomoris
3
150
Chart.js が簡単に使えるようになっていたので OGP 画像生成に使った話
kamekyame
0
170
Rancherの紹介&Update情報(RancherJP Online Meetup #09)
yoshiyuki_kono
0
130
トークン数だけでは測れない — Claude Code 組織展開の効果検証から学んだこと
makikub
0
140
AmazonRoute 53ではじめてのドメイン取得!HTTPS化までの道のりを整理してみた
usanchuu
3
100
Featured
See All Featured
Connecting the Dots Between Site Speed, User Experience & Your Business [WebExpo 2025]
tammyeverts
11
940
The Web Performance Landscape in 2024 [PerfNow 2024]
tammyeverts
12
1.2k
コードの90%をAIが書く世界で何が待っているのか / What awaits us in a world where 90% of the code is written by AI
rkaga
62
44k
The AI Revolution Will Not Be Monopolized: How open-source beats economies of scale, even for LLMs
inesmontani
PRO
3
3.5k
Game over? The fight for quality and originality in the time of robots
wayneb77
1
190
Building a A Zero-Code AI SEO Workflow
portentint
PRO
0
560
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
49
10k
How to Align SEO within the Product Triangle To Get Buy-In & Support - #RIMC
aleyda
2
1.5k
XXLCSS - How to scale CSS and keep your sanity
sugarenia
250
1.3M
Optimising Largest Contentful Paint
csswizardry
37
3.7k
Building the Perfect Custom Keyboard
takai
2
790
エンジニアに許された特別な時間の終わり
watany
107
250k
Transcript
จݙհʢʣ Improving Machine Translation Quality Estimation with Neural Networks Features
Zhiming Chen, Yiming Tan, Chenlin Zhang, Qingyu Xiang, Lilin Zhang. WMT17, pages 551-555 Ԭٕज़Պֶେֶɹࣗવݴޠॲཧݚڀࣨɹ૬ాɹଠҰ
"CTUSBDU w 'FBUVSFFYUSBDUJPO w 2VBMJUZ&TUJNBUJPOʹ͓͍ͯॏཁͳׂΛՌͨ͢ w /FVSBM/FUXPSLͷಛྔΛ2VBMJUZ&TUJNBUJPOʹ༻͍Δ w &NCFEEJOH w
$SPTT&OUSPQZ w 8.5ͷ2VBMJUZ&TUJNBUJPOͰਓखධՁͱͷ૬ؔΛ্ !2
/FVSBM/FUXPSLGFBUVSFT w &NCFEEJOHGFBUVSFT w ೖྗจग़ྗจͷதʹ͓͚Δ֤୯ޠͷຒΊࠐΈදݱΛநग़ w ฏۉΛऔΓɺจͷຒΊࠐΈදݱͱ͢Δ w ೖྗจɺग़ྗจͷຒΊࠐΈදݱΛDPODBUͯ͠ɺಛྔΛಘΔ
!3
/FVSBM/FUXPSLGFBUVSFT w $SPTT&OUSPQZGFBUVSFT w &NCFEEJOHGFBUVSFͰจ຺ͷ͕ؔߟྀ͞Εͳ͍ w ࿈ଓੑɺهԱੑΛ࣋ͭ3//-.͔Β$SPTT&OUSPQZΛநग़ !4
&YQFSJNFOU w ݴޠϖΞɿ&OHMJTI(FSNBO w 8PSEFNCFEEJOH 3//-.ͷUSBJO w &VSPQBSMW $PNNPO$SBXMDPSQVT
BOE/FXT$PNNFOUBSZW W ର༁ͷϖΞ 5SBJO %FW 5FTU &O%F %F&O !5
&YQFSJNFOU w ಛྔΛબʢΈ߹Θͤߟྀʣ w #BTFMJOF จͷ͞ɾɺ-.ͷ֬ͳͲछྨ w 8PSE7FD&NCFEEJOH w
3//-.$SPTT&OUSPQZ w நग़ͨ͠ಛྔΛ4VQQPSU7FDUPS3FHSFTTJPOʹ༻͍Δ w 2&Λߦ͍ɺਓखධՁͨ͠ςετηοτ)5&3ͱͷ%FMUB"WH ૬ؔ Ѝ Л ."& .FBO"WFSBHF&SSPS 3.4& 3PPU.FBO4RVBSF&SSPS ΛௐΔ !6
3FTVMU ߴ͍΄Ͳ˕ ͍΄Ͳ˕ !7
3FTVMU 8PSEWFD୯ମͰ #BTFMJOFΛ্ճͬͨ !8
3FTVMU 8PSEWFDΛج४ʹ͢Δͱ #BTFMJOFΛՃ͑ΔΑΓɺ 3//-.ΛՃ͑ͨํ͕ྑ͘ͳͬͨ !9
3FTVMU 8PSEWFD 3//-.ʹ #BTFMJOFΛՃ͑Δͱɺ είΞ͕͞Βʹྑ͘ͳͬͨ !10
$PODMVTJPO w /FVSBM/FUXPSLͷಛྔΛ2VBMJUZ&TUJNBUJPOʹద༻ w 8PSEWFD&NCFEEJOHGFBUVSFT w 3//-.$SPTT&OUSPQZGFBUVSFT w ࣮ݧͰ֤ಛྔͷ༗༻ੑΛ֬ೝ w
/FVSBM/FUXPSLͷಛྔͰ͋ΔͨΊɺݴޠʹґଘ͠ͳ͍ͷ͕ڧΈ !11