X.Shen | Jason (@hyper_air) profile picture

X.Shen | Jason

@hyper_air

Machine Learning Scientist @Amazon | graduate researcher @CarnegieMellon. Specialized in Recommender systems |Generative AI | Personalization

 Seattle, WA
linkedin.com/in/sx1992
Joined December 2016
  • Tweets13
  • Following419
  • Followers83
4Media
Photo by X.Shen | Jason (@hyper_air)Photo by X.Shen | Jason (@hyper_air)Photo by X.Shen | Jason (@hyper_air)Photo by X.Shen | Jason (@hyper_air)
Presenting our work accepted by @cikm2023! Curious how a sequential recommender captures users' shopping intents from stochastic shopping behaviors? We present G-STO, a graph regularized stochastic Transformer. Released at @AmazonScience #Transformers #RecommenderSystem #ai
Image from X.Shen | Jason (@hyper_air)
1
3
176
@amazon @AmazonScience 🫶 – at Seattle, WA
19
At the @kdd_news. Join us today at the 19th Intl Workshop on Mining and Learning with Graphs (MLG)! @mlgworkshop
Image from X.Shen | Jason (@hyper_air)
X.Shen | Jason (@hyper_air) profile pictureX.Shen | Jason@hyper_air17 Jul 2023
Our work accepted at KDD’s MLG workshop @kdd_news 🎉 Our work learns customers’ shopping journey into a graph representation while distilling a prior knowledge graph for explicit personalizations. Released at @AmazonScience amazon.science/publications/… #KDD2023 #ai #RecommenderSystem
Topological representation learning for e-commerce shopping behaviors

Topological representation learning for e-commerce shopping behaviors

Learning compact representation from customer shopping behaviors is at the core of web-scale E-commerce recommender systems. At Amazon, we put great efforts into learning embedding of customer...

amazon.science
1
3
391
I am glad that our work got selected! So excited. Thanks for spending time for reviewing our work! @cikm2023 – at Seattle, WA
ACM CIKM 2023 (@cikm2023) profile pictureACM CIKM 2023@cikm20235 Aug 2023
We started sending notifications. Some tracks are doing some final checks but authors should receive their decision soon. #CIKM2023
3
201
Our work accepted at KDD’s MLG workshop @kdd_news 🎉 Our work learns customers’ shopping journey into a graph representation while distilling a prior knowledge graph for explicit personalizations. Released at @AmazonScience amazon.science/publications/… #KDD2023 #ai #RecommenderSystem – at Seattle, WA
Topological representation learning for e-commerce shopping behaviors

Topological representation learning for e-commerce shopping behaviors

Learning compact representation from customer shopping behaviors is at the core of web-scale E-commerce recommender systems. At Amazon, we put great efforts into learning embedding of customer...

amazon.science
2
1
17
2,835
Join us at #KDD2023 for our novel research discovery in recommender system at #Amazon
Image from X.Shen | Jason (@hyper_air)
SIGKDD 2026 (@kdd_news) profile pictureSIGKDD 2026@kdd_news14 Jul 2023
📢 Just released: Paper schedules for #KDD2023! Plan ahead and don't miss a beat. 🔍Research track: kdd.org/kdd2023/research-tra… 📈ADS Track: kdd.org/kdd2023/ads-track-pa… Dive into the conference hype and get a sneak peek by watching their promo videos on YouTube: youtube.com/playlist?list=PL…
Image from SIGKDD 2026 (@kdd_news)
1
1
791
Announcing our accepted paper by #KDD2023! Our #NLP based #RecommenderSystem , leverages text representations to actively capture correlation between user preference and item features Paper: arxiv.org/pdf/2305.13731 Code: github.com/JiachengLi1995/Re… #AI #DeepLearning #MachineLearning – at San Francisco, CA
Image from X.Shen | Jason (@hyper_air)
Image from X.Shen | Jason (@hyper_air)
1
1
1
12
2,692
@kdd_news many thanks! 🎉
138
🎉It’s now officially released at @AmazonScience amazon.science/publications/… – at Seattle, WA
Text is all you need: Learning language representations for sequential recommendation

Text is all you need: Learning language representations for sequential recommendation

Sequential recommendation aims to model dynamic user behavior from historical interactions. Existing methods rely on either explicit item IDs or general textual features for sequence modeling to...

amazon.science
X.Shen | Jason (@hyper_air) profile pictureX.Shen | Jason@hyper_air4 Jul 2023
Announcing our accepted paper by #KDD2023! Our #NLP based #RecommenderSystem , leverages text representations to actively capture correlation between user preference and item features Paper: arxiv.org/pdf/2305.13731 Code: github.com/JiachengLi1995/Re… #AI #DeepLearning #MachineLearning
Image from X.Shen | Jason (@hyper_air)
Image from X.Shen | Jason (@hyper_air)
1
1
8
2,043
NEW: Clinton leads Trump by more than two percentage points in popular vote, per AP: Clinton: 65,788,583 (48.3%) Trump: 62,955,363 (46.2%)
194
2,374
360
3,386
Load more