Goto

Collaborating Authors

nlp cypher


The NLP Cypher

#artificialintelligence

With the great engineering minds at Neural Magic, we're all actively attempting to solve a very difficult problem. How do we get these large models into production without blowing up our hardware or our wallet? We all want the same robust performance with our deep learning models. We want them to be accurate, as light as possible, and fast. So… how do we achieve this?


The NLP Cypher

#artificialintelligence

"The biggest downside for the OpenAI embeddings endpoint is the high costs (about 8,000–600,000 times more expensive than open models on your infrastructure), the high dimensionality of up to 12288 dimensions (making downstream applications slow), and the extreme latency when computing embeddings. This hinders the actual usage of the embeddings for any search applications." FYI: I had previously written about this issue over a year ago and even provided a search engine, it seems now more peeps are on top of this issue.


The NLP Cypher

#artificialintelligence

Even OpenAI is feeling the holiday spirit: they open sourced their photorealistic GLIDE model several days ago. Abhishek maps boring model diagrams to code for building intuition! JellyFish is a library for approximate & phonetic matching of strings. A new summarization evaluation metric called the Shannon Score is proposed. It performs the Shannon Game with a language model.


The NLP Cypher

#artificialintelligence

The Generalist Language Model (GLaM), a trillion weight model that can be trained and served efficiently (in terms of computation and energy use) thanks to sparsity, and achieves competitive performance on multiple few-shot learning tasks. GLaM's performance compares favorably to a dense language model, GPT-3 (175B) with significantly improved learning efficiency across 29 public NLP benchmarks in seven categories, spanning language completion, open-domain question answering, and natural language inference tasks.


The NLP Cypher

#artificialintelligence

Hey … so have you ever deployed a state-of-the-art production level inference server? Don't know how to do it? Well… last week, Michael Benesty dropped a bomb when he published one of the first ever detailed blogs on how to not only deploy a production level inference API but benchmarking some of the most widely used frameworks such as FastAPI and Triton servers and runtime engines such as ONNX runtime (ORT) and TensorRT (TRT). PyTorch Lite Inference Toolkit: works with Hugging Face pipeline. Create graphs with your text data.


The NLP Cypher

#artificialintelligence

The Localization Problem (LP) is a glaring dark cloud hanging over the state of affairs in applied deep learning. And acknowledging this problem, I believe, will enable us make better use of applied AI and expand our knowledge in how the business market will form. Defining LP: There is a limit to how much large centralized language models can generalize at scale given: 1) that different users inherently have varying definitions of ground-truths due to inter-dependencies to their unique real-world environment and 2) depending whether or not model performance is mission-critical. In other words, in certain conditions, in order for a model to be optimized for accuracy for a given user, the model needs to be "localized" to its user's ground truth in their data assuming that a model can't afford to be wrong too many times. Example: Imagine there is a kazillion parameter encoder transformer called Hal9000.


The NLP Cypher

#artificialintelligence

Loads of NLP research and code came in this week. But first… is your location actually hidden?… "PySimpleGUI is a Python package that enables Python programmers of all levels to create GUIs." It currently leverages the fairseq library but authors plan to convert to Hugging Face according to their repo on GitHub. The TrOCR model outperforms the current state-of-the-art models on both printed and handwritten text recognition tasks.


The NLP Cypher

#artificialintelligence

We have a long newsletter this week as many new NLP repos were published as tech nerds return from their Summer vacation. This week I'll add close to 150 new NLP repos to the NLP Index. So stay tuned for this update, it will drop this week. Embeddinghub is a database built for machine learning embeddings. It is built with four goals in mind.


The NLP Cypher

#artificialintelligence

Hey Welcome Back! A flood of EMNLP 2021 papers came in this week so today’s newsletter should be loads of fun! 😋 An article on The Gradient had an interesting take on NLU. It describes how a NNs’…


The NLP Cypher

#artificialintelligence

We have a new CLIP implementation from Max Woolf. It allows for faster experimentation and has some new features like using weighted prompts and using icons for priming the model to improve generation quality. It was released today, try it out! From the maker of Rich library, Will McGugan, Textual is a new project where you can create some amazing apps in terminal. Looks like NLP has arrived for cracking encryption.