KDD 2025 Paper Summary
It has been a while since KDD 2025, finally I have had sometime to finish reading all papers that I interested in and summarize some of my learnings in this post  . My primary focus is ...
. My primary focus is ...
It has been a while since KDD 2025, finally I have had sometime to finish reading all papers that I interested in and summarize some of my learnings in this post  . My primary focus is ...
. My primary focus is ...
前一阵子在得道 APP 上读完了这本《PyTorch 性能与显存优化手册》,感觉是一本很不错的 PyTorch Training 入门读物,很适合刚刚接触这个领域的新手小白来读;同时整本书也提供了一个 PyTorch Training 的优化大纲,可以作为一个引子来扩展去学习更加底层的知识和技术。 这篇帖子主要是我读的时候标记下来的一些知识点,并没有包含书籍里面的全部内容。强烈推荐大家去阅...
User sequence modeling has been a hot topic recently in recommendation system thanks to the advancement of transformer architecture and more powerful hardware. In this blog, I would like to have a ...
In this post, I would like to provide a simple summary on the papers I have read in the first quarter of 2025 and discuss some of my thoughts on recent trend regarding recommendation system. Here i...
In this post, I would like to discuss how to build slack, a very popular realtime messaging application, especially for group messaging (a.k.a channel) in cooperation messaging scenario. Functiona...
In this post, I would provide a quick summarization to some papers from Recsys 2024 which I think is pretty interesting and worth a read. The list of paper is selected based on my personal research...
In my pervious post, I have briefly mentioned about sparse features and how they could be used in recommendation system. In this post, let’s have a deeper look into sparse features, as well as revi...
I have been a fan of online game since my parent bought me a PC when I was in 5th grade. I could still feel the excitement when every Thursday night I rushed home and played DaHuaXiYou 2 with my fr...
Reading open source code has been a recommended approach for software engineers to learn. However, in my past 8 years career, I didn’t do a good job on that. After working in a startup for 1 year, ...
这个是读完这篇综述 Efficient Training of Large Language Models on Distributed Infrastructures - A Survey 之后的一个产出,这篇综述文章针对 LLM 的 training 介绍的已经很详细了,但是同时内容过多也不可能全都学完。这里针对自己整理的一些笔记来列一个之后学习的提纲,这个提纲肯定是非常主观的,推荐大家...