Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Knowledge Distillation for BERT

Avatar for Scatter Lab Inc. Scatter Lab Inc.
April 17, 2020
Research
2.3k
0

Knowledge Distillation for BERT

Avatar for Scatter Lab Inc.

Scatter Lab Inc.

April 17, 2020

More Decks by Scatter Lab Inc.

See All by Scatter Lab Inc.
zeta introduction
Avatar for Scatter Lab Inc. scatterlab
0
2k
SimCLR: A Simple Framework for Contrastive Learning of Visual Representations
Avatar for Scatter Lab Inc. scatterlab
0
4.5k
Adversarial Filters of Dataset Biases
Avatar for Scatter Lab Inc. scatterlab
0
2.3k
Sparse, Dense, and Attentional Representations for Text Retrieval
Avatar for Scatter Lab Inc. scatterlab
0
2.4k
Weight Poisoning Attacks on Pre-trained Models
Avatar for Scatter Lab Inc. scatterlab
0
2.2k
Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
Avatar for Scatter Lab Inc. scatterlab
0
2.6k
Beyond Accuracy: Behavioral Testing of NLP Models with CheckList
Avatar for Scatter Lab Inc. scatterlab
0
2.4k
Open-Retrieval Conversational Question Answering
Avatar for Scatter Lab Inc. scatterlab
0
2.3k
What Can Neural Networks Reason About?
Avatar for Scatter Lab Inc. scatterlab
0
2.3k

Other Decks in Research

See All in Research
SoftMatcha 2: 1兆語規模コーパスの超高速かつ柔らかい検索
Avatar for Masataka YONEDA e869120_sub
7
3.7k
LA-Bench 2025：実験指示から 実行可能手順を生成するためのデータセット/LA-Bench 2025: A Dataset for Generating Executable Experimental Procedures from Experimental Instructions
Avatar for Shota Kato stktu
0
130
IA for theory
Avatar for Gabriel Peyré gpeyre
1
360
Claude Code × autoresearch 実践
Avatar for 数理の弾丸 mathbullet
0
230
NLP colloquium: AI Safety Survey
Avatar for Masahiro Kaneko kanekomasahiro
1
990
AIで最適化を解けるか？
Avatar for MIKIO KUBO mickey_kubo
0
150
さくらインターネット研究所テックトーク2026春、研究開発Gr.25年度成果26年度方針
Avatar for KIKUCHI Shunsuke kikuzo
0
180
Pretrain Where? Investigating How Pretraining Data Diversity Impacts Geospatial Foundation Model Performance
Avatar for SatAI.challenge satai
3
100
Spatial Active Noise Control Based on Sound Field Interpolation Incorporating Physical Constraints
Avatar for NII S. Koyama's Lab skoyamalab
0
160
Unified Audio Source Separation (Defense Slides)
Avatar for Kohei Saijo kohei_1979
1
650
AY 2026 Guide to Academic Writing Using Generative AI - Workshop
Avatar for Kenji Saito ks91
PRO
0
150
GLIM とMegaParticles：正規分布近似の限界とタイトカップリング＆パーティクルフィルタの進展 / GLIM and MegaParticles : Progress of the distribution representation in SLAM
Avatar for koide3 koide3
0
720

Featured

See All Featured
sira's awesome portfolio website redesign presentation
Avatar for ElsiraPls elsirapls
0
340
The MySQL Ecosystem @ GitHub 2015
Avatar for Sam Lambert samlambert
251
13k
Building a A Zero-Code AI SEO Workflow
Avatar for Ian Lurie portentint
PRO
0
680
Exploring anti-patterns in Rails
Avatar for Elle Meredith aemeredith
3
470
RailsConf 2023
Avatar for Aaron Patterson tenderlove
30
1.5k
Are puppies a ranking factor?
Avatar for Jono Alderson jonoalderson
2
3.8k
ReactJS: Keep Simple. Everything can be a component!
Avatar for Pedro Nauck pedronauck
666
130k
SEO for Brand Visibility & Recognition
Avatar for Aleyda Solis aleyda
0
4.7k
Building an army of robots
Avatar for Kyle Neath kneath
306
46k
The World Runs on Bad Software
Avatar for Brandon Keepers bkeepers
PRO
72
12k
How To Stay Up To Date on Web Technology
Avatar for Chris Coyier chriscoyier
790
250k
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
Avatar for Akihiro Kokubo akihiro_kokubo
PRO
73
41k

Transcript

  1. ,OPXMFEHF%JTUJMMBUJPOGPS#&35 ੿਌੤ .BDIJOF-FBSOJOH4PGUXBSF&OHJOFFS

  2. ݾର ݾର  1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO  %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35TNBMMFS GBTUFS DIFBQFSBOEMJHIUFS 

    'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF  %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  3. 1BUJFOU,OPXMFEHF%JTUJMMBUJPO GPS#&35.PEFM$PNQSFTTJPO 1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO

  4. • &./-1"DDFQUFE • .JDSPTPGU%ZOBNJDT"*3FTFBSDI • ୭ୡ#&35঑୷दب • ӝઓ,OPXMFEHF%JTUJMMBUJPOҗ׮ܰѱ1BUJFOUܳ੸ਊೣ • 4PGU5BSHFU݅੉ਊೞ૑ঋҊ׮ܲۨ੉যীࢲب೟णೞب۾ೣ

    1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO 1,%GPS#FSU.PEFM$PNQSFTTJPO

  5. 1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO ೟ण੹ۚ • ࣁо૑MPTTܳઙ೤ೣ • )BSEUBSHFUী؀ೠ$SPTT&OUSPQZ-PTT • 4PGU5BSHFU 5FBDIFSNPEFM੄PVUQVU ী؀ೠ$SPTT

    &OUSPQZ-PTT • ઺рۨ੉যٜ੄<$-4>ష௾੄SFQSFTFOBUJPOٜী؀ೠ OPSNBMJ[FE.FBO4RVBSFE-PTT

  6. ೟ण٣పੌ 1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO • 4UVEFOU7BSJBUJPO#&35@ #&35@ • #&35@ CBTF ীࢲ୐Lѐ੄ۨ੉য۽4UVEFOUܳୡӝച •

    1,%4LJQFWFSZLMBZFSীࢲ೟ण • 1,%-BTUMBTULMBZFSীࢲ೟ण

  7. Ѿҗ 1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO ؘ੉ఠ݆ਸٸੜؽ

  8. Ѿҗ 1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO

  9. Ѿҗ 1BUJFOU,OPXMFEHF%JTUJMMBUJPOGPS#&35.PEFM$PNQSFTTJPO

  10. %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35 TNBMMFS GBTUFS DIFBQFSBOEMJHIUFS %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35TNBMMFS GBTUFS DIFBQFSBOEMJHIUFS

  11. • /FVS*14"DDFQUFE • )VHHJOHGBDF • 4UVEFOUݽ؛ীࢲUPLFOUZQFFNCFEEJOHT QPPMFSMBZFS࢏ઁ %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35TNBMMFS GBTUFS

    DIFBQFSBOEMJHIUFS %JTUJM#&35

  12. %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35TNBMMFS GBTUFS DIFBQFSBOEMJHIUFS ೟ण੹ۚ • -PTT • 4UVEFOU5FBDIFS$SPTT&OUSPQZ •

    .BTLFE-.-PTT • $PTJOF&NCFEEJOH-PTT • 4UVEFOU-BZFSܳпفѐ੄ۨ੉য׼ೞաঀஂೣ

  13. Ѿҗ %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35TNBMMFS GBTUFS DIFBQFSBOEMJHIUFS

  14. Ѿҗ %JTUJM#&35 BEJTUJMMFEWFSTJPOPG#&35TNBMMFS GBTUFS DIFBQFSBOEMJHIUFS

  15. 'BTU#&35 "4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF

  16. • 5FODFOU3IJOP#JSE&MJUF5SBJOJOH1SPHSBNীࢲಎ٬߉ই੘ࢿػ֤ޙ • 4QFFE"DDVSBDZܳઑ੺оמೠ#&35.PEFMઁউ • "EBQUJWF*OGFSFODFооמೠ#&35 4FMG%JTUJMMJOHNFDIBOJTN 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF 'BTU#&35

  17. 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF ೟ण੹ۚ

  18. 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF ೟ण੹ۚ • ೟णࣽࢲ • 1SFUSBJOJOH'JOF5VOJOH4FMGEJTUJMMBUJPO • 4FMG%JTUJMMBUJPOزউ୭࢚ਤ#SBODI৬ೞਤ#SBODI੄$SPTT&OUSPQZ҅࢑ • 6ODFSUBJOUZѐ֛بੑ

    • п-BZFS੄#SBODI߹۽6ODFSUBJOUZ҅࢑ • оࢸ6ODFSUBJOUZоծਸࣻ۾"DDVSBDZо֫׮

  19. Ѿҗ 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF

  20. Ѿҗ 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF

  21. Ѿҗ 'BTU#&35B4FMGEJTUJMMJOH#&35XJUI"EBQUJWF*OGFSFODF5JNF

  22. %ZOB#&35 %ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  23. • )VBXFJ/PBIsT"SL-BC • XJEUI EFQUIߑೱਵ۽"EBQUJWFೠ#&35ઁউ • 'VMM#&35.PEFMXJEUIBEBQUJWF#&35XJEUIEFQUIBEBQUJWF#&35ള۲ %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI %ZOB#&35

  24. %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI ೟ण੹ۚ

  25. %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI ೟ण੹ۚ

  26. %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI ೟ण੹ۚ

  27. %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI ೟ण੹ۚ • 8JEUI೟ण • 5FBDIFS੄4PGU5BSHFUҗ$SPTTFOUSPQZ ੐߬٬ۨ੉য )JEEFO-BZFSח.4& • %BUBBVHNFOUBUJPO੸ਊ

    • %FQUI೟ण • NPE E  N@E ੋۨ੉যܳESPQ • FYNVMUJUQMJFSח  ਸESPQ • MPTTחXJEUI৬э਷ؘXFJHIUח׮ܴ • ݃૑݄ীHSPVOEUSVUIMBCFMਸ$SPTT&OUSPQZܳݡৈࢲGJOFUVOJOH

  28. Ѿҗ %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  29. Ѿҗ %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  30. Ѿҗ %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  31. Ѿҗ %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  32. Ѿҗ %ZOB#&35%ZOBNJD#&35XJUI"EBQUJWF8JEUIBOE%FQUI

  33. хࢎ೤פ׮✌ ୶о૕ޙژחҾӘೠ੼੉੓׮ݶ঱ઁٚইېোۅ୊۽োۅ઱ࣁਃ ੿਌੤ .BDIJOF-FBSOJOH4PGUXBSF&OHJOFFS  &NBJMVLKBF!TDBUUFSMBCDPLS