beyonddream

Overview

Over a period of time, I started collecting technical papers that I would, given time and energy, like to implement an open source version of it. I use this to collect some such papers with the status of implementation for each. These papers have only one constraint and that is they should be simplistic enough to be able to implement within < 5000 LOC. I have also captured whether a paper has been implemented or not and if yes then in which language.

Papers

Papers	Implemented ?	Language	Repo
Automatic Keyword Extraction from Individual Documents by Rose et.al.	Yes	Java	https://github.com/beyonddream/JRAKE

yoavg

Some remarks on Large Language Models

Yoav Goldberg, January 2023

Audience: I assume you heard of chatGPT, maybe played with it a little, and was imressed by it (or tried very hard not to be). And that you also heard that it is "a large language model". And maybe that it "solved natural language understanding". Here is a short personal perspective of my thoughts of this (and similar) models, and where we stand with respect to language understanding.

Intro

Around 2014-2017, right within the rise of neural-network based methods for NLP, I was giving a semi-academic-semi-popsci lecture, revolving around the story that achieving perfect language modeling is equivalent to being as intelligent as a human. Somewhere around the same time I was also asked in an academic panel "what would you do if you were given infinite compute and no need to worry about labour costs" to which I cockily responded "I would train a really huge language model, just to show that it doesn't solve everything!". We

veekaybee

ChatGPT Resources

Context

ChatGPT appeared like an explosion on all my social media timelines in early December 2022. While I keep up with machine learning as an industry, I wasn't focused so much on this particular corner, and all the screenshots seemed like they came out of nowhere. What was this model? How did the chat prompting work? What was the context of OpenAI doing this work and collecting my prompts for training data?

I decided to do a quick investigation. Here's all the information I've found so far. I'm aggregating and synthesizing it as I go, so it's currently changing pretty frequently.

Model Architecture

CrockAgile

Floating Point Arcade

![Twitter Follow][twitter]

A Shady Coin Toss

A shady game master approaches...

Hey there! You look like you would enjoy a good game of chance.

slimsag

Because cross-compiling binaries for Windows is easier than building natively

I want Microsoft to do better, want Windows to be a decent development platform-and yet, I constantly see Microsoft playing the open source game: advertising how open-source and developer friendly they are - only to crush developers under the heel of the corporate behemoth's boot.

The people who work at Microsoft are amazing, kind, talented individuals. This is aimed at the company's leadership, who I feel has on many occassions crushed myself and other developers under. It's a plea for help.

The source of truth for the 'open source' C#, C++, Rust, and other Windows SDKs is proprietary

You probably haven't heard of it before, but if you've ever used win32 API bindings in C#, C++, Rust, or other languages, odds are they were generated from a repository called microsoft/win32metadata.

typesanitizer

Here is a list of resources that I have read either fully, or at least to an extent with which I am comfortable with endorsing them.

I've used these resources in different ways:

• Direct application: Some bits and pieces of advice can be applied very directly.
• Reflection: For organizational things, sometimes I would try to write down how I felt my work environment mirrored, and how it differed from the situation described in some work. Writing things down is a useful forcing function to think clearly.
• 1:1 discussions with my manager: We'd take the first 15-20 minutes of

pervognsen

A traditional table-based DFA implementation looks like this:

uint8_t table[NUM_STATES][256]

uint8_t run(const uint8_t *start, const uint8_t *end, uint8_t state) {
    for (const uint8_t *s = start; s != end; s++)
        state = table[state][*s];
    return state;
}

graninas

On hiring Haskellers

Recently I noticed the number of the same two questions being asked again and again on different Haskell resources. The questions were “How to get a Haskell job” and “Why is it so hard to find Haskellers?” Although these two are coming from the opposite sides of the hiring process, the answer is really just one. There is a single reason, a single core problem that causes difficulties of hiring and being hired in the Haskell community, and we should clearly articulate this problem if we want to increase the Haskell adoption.

We all know that there are many people wishing to get a Haskell job. And a visible increase of Haskell jobs looks like there should be a high demand for Haskellers. The Haskell community has also grown like crazy past years. But still, why is it so difficult to hire and to be hired? Why can’t companies just hire any single person who demonstrates a deep knowledge of Haskell in blog posts, in chats, on forums, and in talks? And why do Haskell companies avoid hirin

ih2502mk

Lecture 1: Introduction to Research — [📝Lecture Notebooks] [▶️Video]
Lecture 2: Introduction to Python — [📝Lecture Notebooks] [▶️Video]
Lecture 3: Introduction to NumPy — [📝Lecture Notebooks] [▶️Video]
Lecture 4: Introduction to pandas — [📝Lecture Notebooks] [▶️Video]
Lecture 5: Plotting Data — [📝Lecture Notebooks] [[▶️Vide

vindarel

Developer experience, libraries, performance… (2021/11)

I'll preface this with three things. 1. I prefer schemes over Common Lisps, and I prefer Racket of the Schemes. 2. There is more to it than the points I raise here. 3. I assume you have no previous experience with Lisp, and don't have a preference for Schemes over Common Lisp. With all that out of the way... I would say Common Lisp/SBCL. Let me explain

1. SBCL Is by far the most common of the CL implementations in 2021. It will be the easiest to find help for, easiest to find videos about, and many major open source CL projects are written using SBCL
2. Download a binary directly from the website http://www.sbcl.org/platform-table.html (even for M1 macs) to get up and running (easy to get started)
3. Great video for setting up Emacs + Slime + Quick Lisp https://www.youtube.com/watch?v=VnWVu8VVDbI

Now as to why Common Lisp over Scheme