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
MixPoet
Search
Zhang Yixiao
April 30, 2020
Research
490
4
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
MixPoet
Zhang Yixiao
April 30, 2020
More Decks by Zhang Yixiao
See All by Zhang Yixiao
CoCon
ldzhangyx
0
440
vq-cpc
ldzhangyx
0
420
diora
ldzhangyx
0
340
drummernet
ldzhangyx
0
300
ON-LSTM
ldzhangyx
0
260
Other Decks in Research
See All in Research
ふとした出会いで生まれたSkillが、 社内利用1位になるまで
mikimhk
22
25k
Vers un système bovin allaitant multiperformant :leviers actionnés et trajectoire de réduction de l'empreinte carbone à travers 4 années d'expérimentation à FERM'INOV
institutdelelevage
PRO
0
140
Top 4セキュリティ会議に一通り論文が採択されるまでの道のりと展望 / Top 4 Security Grand Slam
takaomurakami
0
1.1k
Physical AIでモデリングはどう変わるか/How Physical AI Will Change Modeling
stktu
0
330
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
koide3
0
1k
Model Predictive Control for Uncertain Systems: Robust & Data-Driven Designs
kohlerjohannes
0
210
JPA2026_NetworkTutorial_JunKashihara
junkashihara
0
260
Bilan travail en élevage : première restitution des enquêtes 2025-2026
institutdelelevage
PRO
0
270
実例から見るLLMのマンガ理解:実務VQAタスクによる長期的文脈と視覚情報の定性評価
kzmssk
0
210
GeoAgent勉強会 論文紹介と強い推論モデルが与えるインパクト
satai
5
450
Pesée et pointage par imagerie 3D : PHENO3D entre en phase de test terrains
institutdelelevage
PRO
0
150
超効率化への挑戦:1bit LLMの現状と展望
yumaichikawa
0
950
Featured
See All Featured
Paper Plane
katiecoart
PRO
5
54k
How to make the Groovebox
asonas
2
2.5k
New Earth Scene 8
popppiees
4
2.7k
The untapped power of vector embeddings
frankvandijk
2
2k
The Cult of Friendly URLs
andyhume
79
7.1k
It's Worth the Effort
3n
188
29k
Visual Storytelling: How to be a Superhuman Communicator
reverentgeek
2
740
SEO in 2025: How to Prepare for the Future of Search
ipullrank
3
3.9k
Learning to Love Humans: Emotional Interface Design
aarron
275
41k
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
GraphQLの誤解/rethinking-graphql
sonatard
75
12k
Building a Scalable Design System with Sketch
lauravandoore
464
34k
Transcript
MixPoet: Diverse Poetry Generation via Learning Controllable Mixed Latent Space
ArXiv: 2003.06094v1 Presenter: Yixiao Zhang
Overview • Idea: 诗人经历、历史背景等 => 诗歌风格多样化 • Methods: • semi-supervised
VAE • disentangling latent space to sub-spaces • each sub-space corresponds to one factor conditioning • adversarial training
Introduction • 近年的研究,主要考虑语义连贯、主题相关 • 存在diversity的困扰 • diversity: • 主题间多样性:给定两个topic words,生成不同的诗歌
• 主题内多样性:给定一个topic word,生成不同的诗歌 • * 现有的模型倾向于记住常见pattern
Introduction • 生活经历、历史背景、文学流派 => 影响风格
Introduction • MixPoet: semi-supervised VAE • 将latent space分解为sub-spaces,与影响因子一一对应 • 训练阶段:模型预测无label诗歌的factors
• 测试阶段:指定factor的值,生成风格化的诗歌
Related Work • 诗歌生成模型 (RNNs, Memory Models, etc. ) •
多样性的先前研究: • MRL system: 强化学习,鼓励选用高TF-IDF的词汇 • USPG: 无监督最大化style vector和诗歌的mutual information
Related Work • VAE文本生成/诗歌生成 • Yang et. al, 2018b: 学习context-conditioned
latent variable • Hu et al. 2017: 对生成的诗歌进行对抗训练,增强topic相关性 • CVAE 对话多样性: Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders, ACL 2017 • 本文的对抗:在latent space上做对抗训练
Method • topic keyword: mixture empirical distributions: labeled/ unlabeled
Method: Generator • GRU based model • 是length embedding
Method: Semi-supervised C-VAE • 目的是学习 • 引入z • 由于style与semantics耦合 •
不假设y与z的独立性,而是: • 顺序: w => y => z => x (无y label时)
Method: Semi-supervised C-VAE • then for labeled data: • 估计先验
• 和后验 分别使用一个network计算, recon时最小化KL散度。
Method: Semi-supervised C-VAE • labeled data is too limited •
将y看作另一个latent variable • 估计先验 • 和后验 分别使用一个MLP network计算, recon y时最小化KL散度。
Method: Semi-supervised C-VAE • Total Loss:
Method: Latent Space Mixture • 多个factor时的情形: • 独立性假设:
Method: Latent Space Mixture • How to learn mixed latent
space? • For Isotropic Gaussian Space:
Method: Latent Space Mixture • How to learn mixed latent
space? • For Universal Space: 对于condition: ita是噪声,delta是脉冲函数,c是w, y => 从分布中sample出一个值
Method: Latent Space Mixture • 之后使得discriminator区分这两个z • 估计KL散度: • 其中
就是discriminator
Experiments • factors: • 军旅生涯, 乡村生活, 其他 • 时代繁荣, 时代衰落
• => 6种style
Experiments • Baseline: • Ground Truth • C-VAE • USPG
• MRL: SOTA • fBasic, 监督学习模型
Experiments • 多样性,使用Jaccard Similarity指数评价,越低越好 • 诗歌质量:使用Language Model Score(LMS)评价 • 观察:
• 大多数模型倾向生成重复的短语 • MRL与Basic在intra部分只能生成极其相似的诗歌 • C-VAE情况类似
Experiments • Factor Control Results: • 测试生成的诗歌是否与给定因子类别一致
Experiments • 主观实验
Analysis: Style Mixture
Analysis