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
20250226 NLP colloquium: "SoftMatcha: 10億単語規模コー...
Search
Sponsored
·
Your Podcast. Everywhere. Effortlessly.
Share. Educate. Inspire. Entertain. You do you. We'll handle the rest.
→
Hiroyuki Deguchi
February 26, 2025
Research
800
1
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
20250226 NLP colloquium: "SoftMatcha: 10億単語規模コーパス検索のための柔らかくも高速なパターンマッチャー"
Hiroyuki Deguchi
February 26, 2025
More Decks by Hiroyuki Deguchi
See All by Hiroyuki Deguchi
260624_NLP-colloquium: Hubness
de9uch1
0
110
20240820: Minimum Bayes Risk Decoding for High-Quality Text Generation Beyond High-Probability Text
de9uch1
0
360
サブセット探索を用いた高速なkNNニューラル機械翻訳
de9uch1
0
170
20240226_AAMT-Japio
de9uch1
0
200
Searching for Needles in a Haystack: On the Role of Incidental Bilingualism in PaLM’s Translation Capability
de9uch1
0
160
Paper Reading: Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation
de9uch1
0
220
My Research Environmental Setup
de9uch1
0
340
Nearest Neighbor Machine Translation
de9uch1
0
290
Paper Reading - Dynamic Programming Encoding for Subword Segmentation in Neural Machine Translation
de9uch1
0
310
Other Decks in Research
See All in Research
【ローカルAIに向き合う展示会vol.2】液体時間定数型モジュールを用いた オリジナルの双方向エンコーダーモデルNexteraBERT 推論速度向上検討並びにダウンストリーム評価
rikkabotan7
0
130
Visual SLAM未来予測 / Future Prediction in Visual SLAM
koide3
1
710
【Zozo Research 技術共有会】三次元領域の現在と展望
mickey_0226
3
480
The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
shunk031
4
1.1k
ScoreMatchingRiesz for Automatic Debiased Machine Learning and Policy Path Estimation with an Application to Japanese Monetary Policy Evaluation
masakat0
0
310
CVPR2026論文紹介_VLMにとって良いvision encoderとは何か?Rethinking Model Selection in VLM Through the Lens of Gromov-Wasserstein Distance
kobayashi31
1
170
SAKURAONE:An Open Ethernet-based AI HPC System And Its Observed Workload Dynamicsin a Single-Tenant LLM Development Environment
yuukit
1
460
敵対生成プロンプト同時探索による内省型プロンプト最適化
kinoue_smarthr
0
320
AIで最適化を解けるか?
mickey_kubo
0
140
Fukui Shibiten 39 - AI Art
butchi
0
150
Harness Engineering and Al Agent
kzinmr
3
1.8k
Using our influence and power for patient safety
helenbevan
0
370
Featured
See All Featured
The Director’s Chair: Orchestrating AI for Truly Effective Learning
tmiket
1
220
AI Search: Implications for SEO and How to Move Forward - #ShenzhenSEOConference
aleyda
1
1.3k
Testing 201, or: Great Expectations
jmmastey
46
8.2k
Hiding What from Whom? A Critical Review of the History of Programming languages for Music
tomoyanonymous
3
1k
Making Projects Easy
brettharned
120
6.7k
Pawsitive SEO: Lessons from My Dog (and Many Mistakes) on Thriving as a Consultant in the Age of AI
davidcarrasco
0
190
ピンチをチャンスに:未来をつくるプロダクトロードマップ #pmconf2020
aki_iinuma
128
56k
Understanding Cognitive Biases in Performance Measurement
bluesmoon
32
3k
Joys of Absence: A Defence of Solitary Play
codingconduct
1
420
Why Our Code Smells
bkeepers
PRO
340
58k
The B2B funnel & how to create a winning content strategy
katarinadahlin
PRO
1
420
The Curious Case for Waylosing
cassininazir
1
440
Transcript
None
◼ ⚫ ⚫ ⚫ ⚫ ⚫ ◼ ⚫ ⚫ ⚫
◼ ◼ ◼ ◼
◼ ◼ ◼ ◼
◼ ◼ ◼ ◼
𝑤 ◼ 𝑤 ◼ ⚫ 𝑤 ⚫ 𝑤 ◼ ⚫
◼ ⚫ ◼ ⚫ ◼ ⚫ (Radovanovic+, JMLR2010) ⚫ ▶
Wang+, (arxiv) 2022, “Text Embeddings by Weakly-Supervised Contrastive Pre-training”. Radovanovic+, JMLR 2010, “Hubs in space: Popular nearest neighbors in high-dimensional data”.
◼ ◼ ◼ ◼
◼ ⚫ 𝐩 = 𝑝1 , … , 𝑝𝑀 ∈
Σ∗ ⚫ 𝐭 = 𝑡1 , … , 𝑡𝑁 ∈ Σ∗ ▶ Σ∗ ◼ ⚫ ⚫ ◼ ◼
◼ ⚫ 𝑤 ∈ 𝒱 𝐷 ⚫ 𝐯𝑤 ∈ ℝ𝐷
≔ 𝑤 ⚫ ▶ cos 𝐯person , 𝐯people > cos 𝐯person , 𝐯bird ▶
◼ ◼ ⚫ ※
◼ ⚫ 𝑡𝑖 = 𝑝𝑗 cos 𝐯𝑡𝑖 , 𝐯𝑝𝑗 ≥
𝛼 ▶ 𝛼 = 1.0 ※ 𝛼 = 0.7 ◼ ⚫ ⚫
𝑡1 𝑡2 𝑡3 𝑡4 𝑡5 𝑡6 𝑡7 𝑡8 𝑡9 𝑡10
𝑡11 𝑡12 𝑡13 ◼ ◼ ◼ ⚫
◼ 𝒮𝑤 ≔ 𝑣 ∈ 𝒱 cos 𝐯𝑣 ⊤𝐯𝑤 ≥
𝛼 ⚫ 𝑤 𝒮we 𝒮talk 𝒮about
⇔ 𝒮we , 𝒮talk , 𝒮about ⇔ 𝒮we , 𝒮talk
, 𝒮about 𝑖, 𝑖 + 1, 𝑖 + 2 𝑖 𝒮we 𝒮talk 𝒮about
𝒮we 𝒮talk 𝒮about 𝒮we ℳ ℳ ← 1,10,6 𝒮talk ℳ
ℳ′ ← 2 − 1, 11 − 1,7 − 1 = 1,10,6 ℳ ← ℳ ∩ ℳ′ = 1,10,6 𝒮about ℳ ℳ′ ← 8 − 2, 12 − 2 = 6,10 ℳ ← ℳ ∩ ℳ′ = 6,10 ℳ
𝒮we 𝒮talk 𝒮about 𝑡1 𝑡2 𝑡3 𝑡4 𝑡5 𝑡6 𝑡7
𝑡8 𝑡9 𝑡10 𝑡11 𝑡12 𝑡13
𝒮𝑝1 𝐼𝒮𝑝1 𝒮𝑝𝑀 𝐼𝒮𝑝𝑀 𝒮𝑝1 𝐼𝒮𝑝1 ℳ ℳ ← 𝐼𝒮𝑝1
𝑘 = 2, … , 𝑀 ℳ′ ← 𝑖 − 𝑘 + 1 𝑖 ∈ 𝐼𝒮𝑝𝑘 ℳ ← ℳ ∩ ℳ′ ℳ 𝐩 = 𝑝1 , … , 𝑝𝑀
◼ ⚫ ▶ ▶ ⚫
◼ ◼ ⚫ (Wang+, 2024) ⚫ (Douze+, 2024) (Malkov &
Yashunin, IEEE TPAMI, 2018) ◼ ⚫ ▶ 𝛼 = 0.55 (Pennington+, EMNLP2014) ▶ 𝛼 = 0.50 (Grave+, arXiv:1802.06893) Wang+, arXiv:2402.05672, “Multilingual E5 Text Embeddings: A Technical Report”. Douze+, arXiv:2401.08281, “The Faiss library”. Malkov & Yashunin, IEEE TPAMI, 2018, “Efficient and Robust Approximate Nearest Neighbor Search Using Hierarchical Navigable Small World Graphs”. Pennington+, EMNLP2014, “GloVe: Global Vectors for Word Representation”. Grave+, arXiv:1802.06893, “Learning Word Vectors for 157 Languages”.
◼ ⚫ ▶ ⚫
◼ ⚫ ⚫ ◼ ⚫
◼ ⚫ (Crane, IJDL 2023) ⚫ (Bothwell+, EMNLP2023) ◼ Crane,
IJDL 2023, “The Perseus Digital Library and the future of libraries.”. Bothwell+, EMNLP2023, “Introducing Rhetorical Parallelism Detection: A New Task with Datasets, Metrics, and Baselines”.
◼ ⚫ ⚫ ⚫ ◼ ◼
◼ ⚫ ◼ ⚫ ▶ ◼ ⚫ ▶ ▶ ⚫
⚫
◼ ⚫ 𝐼𝒮𝑝𝑘 ▶ ⚫ ◼ ⚫
◼ ◼ ⚫ ⚫ ⚫ 𝑂 1 ▶ ◼ ⚫
⚫ ⚫ 𝑂 log |𝐵| ▶
◼ ⚫ ⚫ ◼ ⚫ ⚫ ▶ ⚫