@seeyoojm Profile picture

Jaemin Yoo

@seeyoojm

Assistant Professor at KAIST EE & AI. Former Postdoc @CarnegieMellon. Data Mining, Machine Learning, Graph Neural Networks, and Anomaly Detection

Similar User
Dongkwan Kim photo

@_dongkwan_kim

Jinheon Baek photo

@jinheonbaek

Hoyeon Chang photo

@hoyeon_chang

Hyeon min Yun photo

@hyeonmin_Lona

Jay Shin photo

@jshin491

Sangmin Bae photo

@raymin0223

Sungnyun Kim photo

@kim_sungnyun

Sohee Yang @ ACL 2024 photo

@soheeyang_

Seokeon Choi photo

@SeokeonC

JinYeong Bak photo

@NoSyu

Jaehong Yoon photo

@jaeh0ng_yoon

Minki Kang photo

@mkkang_1133

Minguk_Kang photo

@minguk_kang

Sihyun Yu photo

@sihyun_yu

Sungdong Kim photo

@SungdongKim4

Jaemin Yoo Reposted

This is the system prompt for Apple Intelligence. Turns out Apple's prompt engineers are as clueless about how LLM work as all the others.

Tweet Image 1

Jaemin Yoo Reposted

This LLMs on Graphs survey is accepted to the TKDE journal with 700+⭐️ in our resource repo. We have updated the survey to include more discussions on the future directions. Feel free to help a look!

📢We have finalized our 𝐋𝐋𝐌𝐬 𝐨𝐧 𝐆𝐫𝐚𝐩𝐡𝐬 survey by adding more insightful discussions. If you are interested in LLMs on structure data, don't miss this paper (with a resource repo)! #Graph #LLM paper:arxiv.org/abs/2312.02783 resource:github.com/PeterGriffinJi…

Tweet Image 1


Jaemin Yoo Reposted

here's my prompt of the day: """ You are an extremely thoughtful and verbose assistant who can help user extremely well with complicated problems that require sophisticated reasoning. You are mathematically enlightened and rely heavily on mathematical and statistical reasoning.…


Jaemin Yoo Reposted

PyTorch dataloader has two terrible default settings. Fixing it gave me ~5x speedup ⚡️ In any standard PyTorch model training with GPU: - .to(device) transfers the data to the GPU. - Everything executes after this happens on the GPU. This means when the GPU is working, the CPU…

Tweet Image 1

Jaemin Yoo Reposted

MIT researchers propose using LLMs for anomaly detection in time-series data. Leveraging their autoregressive capabilities to handle sequential data without fine-tuning. 📌 The SigLLM framework transforms time-series data into text-based inputs for LLM processing, enabling…

Tweet Image 1

Jaemin Yoo Reposted

arXiv -> alphaXiv Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. @askalphaxiv You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!


Jaemin Yoo Reposted

after months of writing, revising, reorganizing and partially translating the contents, you can now access the completed textbook at kyunghyuncho.me/linear-algebra…. this book comes in two languages; English and Korean (or more like Korean-English mix). so, click the link, and download…

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Jaemin Yoo Reposted

READ: 130-page cheat sheet outlining key #MachineLearning concepts: bit.ly/130pgMLsheet

Tweet Image 1

Jaemin Yoo Reposted

One month is still left to participate in the 2nd ICML Topological Deep Learning Challenge! 🚀 Implement a topological lifting and get published 💫 Please check out our website for the full details 📢: pyt-team.github.io/packs/challeng…

Tweet Image 1

Jaemin Yoo Reposted

지난 몇 달간 여러 AI 챗봇 서비스를 방황하다 결국 @poe_platform 에 정착했다. 단 하나의 구독으로 ($20/mo) 모든 LLM과 대화할 수 있는데, 채팅 세션 중 "@"로 다른 LLM을 태그 시키 대화 참여시킬 수 있어 답변을 교차 검증하는게 정말 편하다. 심심할땐 LLM끼리 서로 랩배틀도 시킬 수 있다.

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Jaemin Yoo Reposted

Graph Machine Learning course Lecture 5 presents Recommendation on Graphs 🌟 Google PageRank, collaborative & content recommendation, low-rank matrix completion, graph diffusion filtering, non-negative matrix factorization are the prominent techniques. storage.googleapis.com/xavierbresson/…

Tweet Image 1
Tweet Image 2
Tweet Image 3
Tweet Image 4

Jaemin Yoo Reposted

인공지능을 위한 기초 수학과 관련된 포스팅을 할까 싶긴 한데 다른 분들도 많이 잘 적어두셔서... mml-book.github.io 기초 수학이 필요하시다면 이 책을 읽으시면 됩니다. 이게 전부는 아니지만 필요한건 있어요.


Jaemin Yoo Reposted

Data Augmentation is a Hyperparameter: Cherry-picked Self-Supervision for Unsupervised Anomaly De... Jaemin Yoo, Tiancheng Zhao, Leman Akoglu. Action editor: Chang Xu. openreview.net/forum?id=HyzCu… #supervised #anomaly #detection


Jaemin Yoo Reposted

This 10-page data science cheat-sheet covers machine learning, probability, statistics & more: bit.ly/2mmrIMB (@MaverickLin v/@KirkDBorne)

Tweet Image 1

Jaemin Yoo Reposted

Is this going to be on the test?

Tweet Image 1

Jaemin Yoo Reposted

지난 주말에 SUSC에서 발표한 "대학생 때 알았더라면 좋았을 것들 (2023 Ver)" 자료를 공유드립니다. 작년에 고려대학교에서 발표했던 자료에서 몇 가지 내용을 추가해 갱신했습니다. 컴퓨터 공학을 공부하고 있는 대학생분들께 도움이 되었으면 합니다. 감사합니다. 🤗🤗🤗 present.do/documents/64b5…


Jaemin Yoo Reposted

9 Distance Measures in data science with algorithms. via: @gp_pulipaka

Tweet Image 1

Jaemin Yoo Reposted

Probabilistic Machine Learning by @sirbayes is the bible of machine learning and in its latest edition also has a chapter on conformal prediction. And it is available for free online. probml.github.io/pml-book/book2… #machinelearning #textbooks

Tweet Image 1

Jaemin Yoo Reposted

파이썬 3.11은 어떻게 빨라질 수 있었을까? yozm.wishket.com/magazine/detai… 이 글에서는 파이썬 3.11이 어떻게 빨라졌는지에 대해 알아보려고 한다. - 더 빠른 파이썬을 위한 파이썬 3.11 - 파이썬 3.11의 주된 변경점 #파이썬 #개발 #개발자

Tweet Image 1

Jaemin Yoo Reposted

Students looking to boost their research skills can find immense value in @jbhuang0604's insightful tips. A treasure trove of knowledge awaits you here: github.com/jbhuang0604/aw…


Loading...

Something went wrong.


Something went wrong.