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
意味の解析
Search
Yumeto Inaoka
February 21, 2017
Technology
0
260
意味の解析
2017/02/22のB3ゼミで発表
Yumeto Inaoka
February 21, 2017
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
120
文献紹介: Open Domain Web Keyphrase Extraction Beyond Language Modeling
yumeto
0
160
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
120
文献紹介: PAWS: Paraphrase Adversaries from Word Scrambling
yumeto
0
88
文献紹介: Beyond BLEU: Training Neural Machine Translation with Semantic Similarity
yumeto
0
210
文献紹介: EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing
yumeto
0
260
文献紹介: Decomposable Neural Paraphrase Generation
yumeto
0
180
文献紹介: Analyzing the Limitations of Cross-lingual Word Embedding Mappings
yumeto
0
170
Other Decks in Technology
See All in Technology
[RSJ24] Object Segmentation from Open-Vocabulary Manipulation Instructions Based on Optimal Transport Polygon Matching with Foundation Models
keio_smilab
PRO
0
130
Functional TypeScript
naoya
10
3.9k
HolidayJp.jl を作りました
mrkn
0
120
Javaにおける関数型プログラミンへの取り組み
skrb
7
280
なぜクラウドサービスで Web コンソールを提供するのか
shuta13
4
2k
すぐに始めるAWSコスト削減。短期でできる改善策と長期的な運用負荷軽減への取り組み方を解説
ncdc
1
520
ビジネスとエンジニアリングを繋ぐプロダクトを中心とした組織づくりの実践
sansantech
PRO
1
130
LLMに日本語テキストを学習させる意義
ksaito
13
3.6k
[RSJ24] Object Retrieval in Large-Scale Indoor Environments Using Dense Text with a Multi-Modal Large Language Model
keio_smilab
PRO
0
230
Eventual Detection Engineering
ken5scal
0
970
EitherT_with_Future
aoiroaoino
1
920
疎通2024
sadnessojisan
5
870
Featured
See All Featured
Imperfection Machines: The Place of Print at Facebook
scottboms
263
13k
Pencils Down: Stop Designing & Start Developing
hursman
118
11k
The Psychology of Web Performance [Beyond Tellerrand 2023]
tammyeverts
34
1.9k
GitHub's CSS Performance
jonrohan
1029
450k
Why Our Code Smells
bkeepers
PRO
334
56k
How to Ace a Technical Interview
jacobian
275
23k
Music & Morning Musume
bryan
46
6k
What’s in a name? Adding method to the madness
productmarketing
PRO
21
3k
The Straight Up "How To Draw Better" Workshop
denniskardys
230
130k
[RailsConf 2023] Rails as a piece of cake
palkan
45
4.6k
The Illustrated Children's Guide to Kubernetes
chrisshort
46
48k
実際に使うSQLの書き方 徹底解説 / pgcon21j-tutorial
soudai
166
48k
Transcript
ҙຯͷղੳ #θϛ ࣗવݴޠॲཧݚڀࣨɹҴԬເਓ
ࢀߟʹͨ͠ॻ੶
ൃද༰ ˗ ҙຯͷղੳͷඞཁੑ ˗ ड़ޠ߲ߏ ˗ ֨ͱ֨ϑϨʔϜ ˗ ֨ϑϨʔϜͷࣗಈߏங
ҙຯͷղੳͷඞཁੑ ˗ ߏจղੳͰจͷߏΛղ໌ ˗ จͷҙຯ8)ͰදݱՄೳ ˗ จͷߏ8)Λ໌ࣔ͠ͳ͍ ɹˠ͞ΒͳΔॲཧ͕ඞཁ
ߏͱҙຯͷҧ͍ " +PIOCSPLFUIFXJOEPXXJUIBIBNNFS # 5IFXJOEPXCSPLF $ "IBNNFSCSPLFUIFXJOEPX ˗ ҙຯ΄΅ಉ͕ͩ͡ߏશવҧ͏ ˗
ओޠతޠΛٻΊΒΕͯҙຯෆ໌
ߏͱҙຯͷҧ͍ " ళ͕ΩϟϯϖʔϯͰച্Λഒ૿ͨ͠ɻ # ച্͕ഒ૿ͨ͠ɻ $ Ωϟϯϖʔϯ͕ച্Λഒ૿ͨ͠ɻ ˗ ओޠҧ͏͕ҙຯಉ͡ ˗
ޠಈࢺͷ߹ʹੜ͡Δ
ड़ޠ߲ߏ ˗ ड़ޠΛҙຯͷத৺ͱ͢Δ ˗ ड़ޠΛิ͏߲ͱͷؔͰҙຯΛଊ͑Δ
֨ ˗ ߲ͷׂ ˗ ද֨ɿදతʹܾ·Δ֨ ˗ ਂ֨ɿਂతɾҙຯతʹܾ·Δ֨
ද֨ ओ֨ ओޠͷׂ ର֨ తޠͷׂ ༩֨ ؒతޠͷׂ
ຊޠͷද֨ ˗ ຊޠޠॱ͕ࣗ༝ ˗ ޠॱ͔Β͕֨۠ผͰ͖ͳ͍ ˗ ֨ॿࢺ͕֨Λࣝผ ˗ ֨ॿࢺʹରԠͨ͠Ψ֨ɾϮ֨ɾχ֨ͳͲ
ਂ֨ ಈ࡞ओ֨ ಈ࡞ΛҾ͖ى͜͢ऀ ܦݧऀ֨ ৺ཧݱΛମݧ͢Δऀ ಓ۩֨ ग़དྷࣄͷݪҼ ର֨ ҠಈมԽͷର
ݯઘ֨ ରͷҠಈʹ͓͚Δى ඪ֨ ରͷҠಈʹ͓͚Δऴ ॴ֨ ग़དྷࣄ͕ى͜ΔॴҐஔ ࣌ؒ֨ ग़དྷࣄ͕ى͜Δ࣌ؒ
֨ϑϨʔϜ " +PIOCSPLFUIFXJOEPXXJUIBIBNNFS # 5IFXJOEPXCSPLF $ "IBNNFSCSPLFUIFXJOEPX { CSFBL
+PIOಈ࡞ओ֨ XJOEPXର֨ IBNNFSಓ۩֨ ֨ϑϨʔϜ
ຊޠͷ֨ղੳ ΦʔϒϯͰϋϜͤͨύϯΛম͍ͨɻ ˗ ʮϋϜΛͤΔʯͷͰϋϜϮ֨ ˗ ʮύϯʹͤΔʯͷͰύϯχ֨ ˠ ඞͣ֨͠ॿࢺࣔ͞Εͳ͍ ˠ ࿈ମम০અ໌֬Ͱͳ͍
֨ϑϨʔϜͷࣗಈߏங ˗ ड़ޠɾ֨ɾ߲ͷΈ߹ΘͤΛίʔύε ͔Βࣗಈߏங ˗ ສจنͷऍ༩ίʔύεͰ ߴਫ਼ͳղੳʹඞཁͳྔͱͯ͠ෆे ˗ େྔͷੜίʔύε͔Βࣗಈߏங
֨ϑϨʔϜͷࣗಈߏங ˗ ߏจղੳͷ݁Ռ͔Βड़ޠ߲ߏΛநग़ ˗ ߏจతᐆດੑͷͳ͍ͷ͚ͩΛநग़ ˗ ड़ޠͷલͷ߲༻͍ͯநग़͢Δ
֨ϑϨʔϜͷࣗಈߏங ΦʔϒϯͰϋϜͤͨύϯΛম͍ͨɻ ˗ ΦʔϒϯͰͤͨʁ ˗ ΦʔϒϯͰম͍ͨʁ ˗ ߏจʹᐆດੑͷ͋Δͷແࢹ
֨ϑϨʔϜͷࣗಈߏங ˗ ߏจղੳͷ݁Ռ͔Βड़ޠ߲ߏΛநग़ ˗ ߏจతᐆດੑͷͳ͍ͷ͚ͩΛநग़ ˗ ड़ޠͷલͷ߲༻͍ͯநग़͢Δ
֨ϑϨʔϜͷࣗಈߏங " ΦʔϒϯͰϋϜͤͨύϯΛম͍ͨ # ࣇಐʹઌੜ͕खΛম͍ͨ $ ൴$%ʹσʔλΛম͍ͨ ˗ ಉ͡ʮম͍ͨʯ͕ͩҙຯҧ͏ ˗
લͷ߲Ληοτʹͯ͠நग़
֨ϑϨʔϜࣗಈߏஙͷྫ ژେֶ֨ϑϨʔϜ ˗ ԯຊޠจΛ༻ ˗ ສޠͷड़ޠ ˗
$16ͷΫϥελܭࢉػͰिؒ ࠇڮɾՏݪݚڀࣨʮژେֶ֨ϑϨʔϜʯ IUUQOMQJTUJLZPUPVBDKQJOEFYQIQ ژେֶ֨ϑϨʔϜ
ൃදͨ͠༰ ˗ ҙຯͷղੳͷඞཁੑ ˗ ड़ޠ߲ߏ ˗ ֨ ˗ ֨ϑϨʔϜͷࣗಈߏங