Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Sign up for free
Menu
Search
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Features
All features
Private URLs
Password Protection
Custom URLS
Scheduled publishing
Remove Branding
Restrict embedding
Deck Collections
Notes
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Explore
Featured decks
Featured speakers
Programming
Technology
Storyboards
Pricing
Search
Sign in
Sign up for free
文献紹介: Attention is not Explanation
Search
Yumeto Inaoka
March 19, 2019
Research
530
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
文献紹介: Attention is not Explanation
Yumeto Inaoka
March 19, 2019
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
230
文献紹介: Open Domain Web Keyphrase Extraction Beyond Language Modeling
yumeto
0
290
文献紹介: Self-Supervised_Neural_Machine_Translation
yumeto
0
200
文献紹介: Comparing and Developing Tools to Measure the Readability of Domain-Specific Texts
yumeto
0
210
文献紹介: PAWS: Paraphrase Adversaries from Word Scrambling
yumeto
0
220
文献紹介: Beyond BLEU: Training Neural Machine Translation with Semantic Similarity
yumeto
0
330
文献紹介: EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing
yumeto
0
440
文献紹介: Decomposable Neural Paraphrase Generation
yumeto
0
260
文献紹介: Analyzing the Limitations of Cross-lingual Word Embedding Mappings
yumeto
0
300
Other Decks in Research
See All in Research
「AIとWhyを深堀る」をAIと深堀る
iflection
0
610
長時間動画QAにおけるマルチエージェント推論 ・SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent Collaboration
murakawatakuya
1
200
Anthropic が提案する LLM の内部状態を自然言語で説明可能にした Natural Language Autoencoders / Natural Language Autoencoders Produce Unsupervised Explanations of LLM Activations
shunk031
0
220
JICA QUEST 共創×革新プログラム Impact Report(海ノ向こうコーヒー)
ontheslope
0
590
論文紹介: Understanding Epistemic Language with a Language-augmented Bayesian Theory of Mind
hisaokatsumi
0
160
IA for theory
gpeyre
1
410
重要だけど測れていないもの:高齢者ケアの見えない課題
theoriatec2024
0
520
PHTalks Bengaluru - SSRF When All Else Fails
dk999
0
1.1k
Fukui Shibiten 39 - AI Art
butchi
0
190
Spatial Active Noise Control Based onSound Field Interpolation Incorporating Physical Constraints
skoyamalab
0
190
[IR Reading 2026春 論文紹介] LLM-based Listwise Reranking under the Effect of Positional Bias (ECIR 2026) /IR-Reading-2026-Spring
koheishinden
PRO
0
410
Visual SLAM未来予測 / Future Prediction in Visual SLAM
koide3
1
1k
Featured
See All Featured
How to audit for AI Accessibility on your Front & Back End
davetheseo
0
530
The Myth of the Modular Monolith - Day 2 Keynote - Rails World 2024
eileencodes
28
3.6k
Become a Pro
speakerdeck
PRO
31
6.2k
Large-scale JavaScript Application Architecture
addyosmani
515
110k
How STYLIGHT went responsive
nonsquared
100
6.3k
WCS-LA-2024
lcolladotor
0
830
Lightning Talk: Beautiful Slides for Beginners
inesmontani
PRO
2
680
A Modern Web Designer's Workflow
chriscoyier
699
190k
Fashionably flexible responsive web design (full day workshop)
malarkey
408
67k
Stewardship and Sustainability of Urban and Community Forests
pwiseman
0
520
Highjacked: Video Game Concept Design
rkendrick25
PRO
1
450
Leo the Paperboy
mayatellez
9
2.3k
Transcript
Attention is not Explanation 文献紹介 2019/03/19 長岡技術科学大学 自然言語処理研究室 稲岡 夢人
Literature 2 Title Attention is not Explanation Author Sarthak Jain,
Byron C. Wallace Conference NAACL-HLT 2019 Paper https://arxiv.org/abs/1902.10186
Abstract Attentionは入力単位で重みを示す その分布が重要性を示すものとして扱われることがある → 重みと出力の間の関係は明らかにされていない 標準的なAttentionは意味のある説明を提供しておらず、 そのように扱われるべきでないことを示す
3
調査方法 1. 重みが重要度と相関しているか 2. 事実と反する重み設定が予測を変化させるか 4
Tasks Binary Text Classification Question Answering (QA)
Natural Language Inference (NLI) 5
Datasets 6
Results 7
Definitions 出力結果の比較に使用する距離 Attentionの比較に使用する距離 8
Results Attentionの変化を大きくしても結果が変化しない → 出力への影響が小さい 9
Results DiabetesはPositiveのクラスにおいては影響が大きい → 高精度で糖尿病を示すトークンが存在するため 10
Adversarial Attention 出力を大きく変化させるようにAttentionを変化させる Attentionが少し変化しただけで出力が大きく変化するか ← Attentionの挙動を確認 12
Results 少しのAttentionの変化で出力が大きく変化している 13
Conclusions 重要度とAttentionの重みは相関が弱い 事実に反する重みは必ずしも出力を変化させない Seq2seqについては今後の課題とする 14