马东锡NLP读书笔记
Sverige_ Dong-seok一个NLP研究员的读书笔记
| Platform |  | Pricing | Only free issues | Publishes | Daily |
|---|---|---|---|---|---|
| Issues | 9 | Founded | 3 years ago | Last Issue | 3 years ago |
| Active | |||||
Read this Newsletter
sverigedongseok.substack.comLatest Issues
[Formers]大语言模型是如何发展到被prompting就可以完成NLP任务的，prompt到底是什么？
本条读书笔记以从pretraining谈起，简要介绍LLM从GPT-1, BERT, BART, 到GPT-2的发展，并配合简述我最喜欢的NLP研究员Timo Schick(Toolfomer的作者)的三篇prompt-based learning的文章简要介绍prompt的演变历程。
Pretraining
Pretraining最早应用在CV领域，如ImageNet。 思想是通过预训练，使得深度神经网络习得某类型数据（视觉或文字）普遍性的feature，然后运用到...
[CoT]关于AutoGPT的疑惑, 它到底有没有CoT?
昨天发了关于AutoGPT的thread，断言它的prompt没有CoT是它的致命缺陷。 今日重新思考了这个问题，并在两位读者的指正和讨论下，重新看了AutoGPT的代码，整理了一下思考，感觉昨天的结论下的有些不严谨。 那么AutoGPT到底有没有CoT?
根据AutoGPT的代码，粗略地画了它构建prompt message的流程图。 prompting llm是AutoGPT的关键，所以整个过程中它到底是怎样构建prompt message非常重要。 总的来说，四个部分...
[CoT]从AutoGPT谈起，看大语言模型使用工具的发展
这两天，推上最热话题莫过AutoGPT。 AutoGPT的目的是让大语言模型使用外部工具（例如google）完成复杂任务，这与我最近读的论文话题一致。 此读书笔记，从AutoGPT的源码出发谈论它的局限性，以及简短地总结几篇文章来看LLM使用工具的发展。
抛开炫酷的demo，AutoGPT最核心的代码，是prompt message的构成方式，即： 把用户输入的ai name/role/goals直接合并在它默认的prompt message中。 它的本质，还是在promp...
[CoT] ReAct: 大语言模型推理，决断和行动的关键
最近大语言模型突破了文字处理任务的限制，向智能coordinator的角色转化。 一个疑问随之而来，“LLM到底如何决断并采取行动来调用不同的api的？” 本条post读书笔记，通过解读论文ReAct，同时介绍langchain的一个具体例子来试图回答这个问题。
paper: ReAct: Synergizing Reasoning and Acting in Language Models. 关键词 [推理]，[行动] LLM有没有主动推理能力？目前没有确切答案。 但是可...
[CoT]从Chain-of-Thought (CoT) 到 Let's think step by step
Paper 1: Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
此文是CoT的开山之作，作者Jason Wei当时就职于google brain，如今在openAI，同时也是ChatGPT的重要作者。 CoT是解锁LLM 推理能力的重要钥匙，一举开启了挖掘LLM隐藏技能的新的paradigm.
在CoT出现之前, LLM的发展遇到了尴尬的瓶颈，模型越来越大，处理文字能力越来越强...
Key Facts
Audience Metrics
Subscribers, engagement, traffic and sponsorship for 马东锡NLP读书笔记.
| Subscribers | Monthly Web Visits | Accepts Sponsors | |||
|---|---|---|---|---|---|
| Estimated Cost per Ad | |||||
Explore More
Authors
The writers behind this newsletter.
Swedish citizen, 三流 PhD in NLP
Frequently Asked Questions
How can I access the email archive for 马东锡NLP读书笔记?
You can find recent issues that have been published by 马东锡NLP读书笔记 on Reletter by scrolling up to where it says Latest Issues. Tap on the link for any of the most recent emails or hit More Issues to see older ones.
How many subscribers does 马东锡NLP读书笔记 have?
To see how many people subscribe to 马东锡NLP读书笔记, simply upgrade your Reletter account. We provide readership numbers and lots of other stats for this newsletter so you can decide if it's worth reaching out to.
How can I advertise in 马东锡NLP读书笔记?
Newsletter advertising can be extremely effective when it's done right. Before you pitch 马东锡NLP读书笔记 as a potential sponsor or partner, make sure that you've done your research and checked its newsletter stats with Reletter.
Then, personalize one of our winning pitching templates and send it to the right person using the contact info provided.
How much does it cost to sponsor a publication like 马东锡NLP读书笔记?
Newsletter ad rates (or CPM) vary depending on many factors, including industry, number of subscribers, open rate, ad placement and more.
To find out how much an ad will cost, contact 马东锡NLP读书笔记 using the contact information provided and ask for a copy of their media kit.
How can I find newsletters related to 马东锡NLP读书笔记?
Scroll up to where it says Related Newsletters to see other publications like 马东锡NLP读书笔记. You can also search our email newsletter directory to discover other newsletters that cover the topics you're interested in.
How do I contact 马东锡NLP读书笔记?
Reletter provides this newsletter's website URL above, where you will often find their contact information. We also provide links to associated social media accounts and pitching templates so you can reach out fast.