Viewing profile — isaacfung

Some models use vector quantized variational autoencoders to discretize images into sequences of discrete symbols from a fixed codebook. https://grok.com/share/bGVnYWN5LWNvcHk%3D_5…

What AI tools have you used recently? Have you verified if they all use models trained on copyrighted material with permission?

Something similar happened to Google Map in Hong Kong recently. A dozen schools had their names changed by some pranksters. It's surprisingly easy. https://www.scmp.com/news/hong-k…

Also this blog post https://yang-song.net/blog/2021/score/

You may find the huggingface course more approachable https://huggingface.co/learn/diffusion-course/en/unit0/1

How do you keep table structure?

It sees embeddings that is trained to encode semantic meanings. The way we tokenize is just a design choice. Character level models(e.g. karpathy's nanoGPT) exist and are used for …

The text is converted to embeddings after tokenization. The neural networwk only sees vectors. Imagine the original question is posed in English but it is translated to Chinese and…

The possibility seems far beyond gaming(given enough computation resources). You can feed it with videos of usage of any software or real world footage recorded by a Go Pro mounted…

When people say "LLMs are not capable of innovation", what exactly do they consider as innovation? If LLMs are not capable of innovation on their own, what if we augment them with …

bro, it's not just about the money. He's promoting himself, building up a portfolio, learning new skills and entertaining himself.

We have translators. Doesn't mean we can't replace them with a cheaper, more accessible tool. That's the whole point of automation. Reasoning stuff is not useless. They provably(ac…

This repo has some less trivial examples. https://github.com/ganarajpr/awesome-dspy You can try STORM(also from Stanford) and see the prompts it generates automatically, it tries t…

Aside from being used alone, T5 is also used as the text encoder of some recent multimodal models. https://stability.ai/news/stable-diffusion-3-research-paper https://t5tts.github.…

I don't get why some people seem to think the only way to use a LLM is for next token prediction or AGI has to be bult using LLM alone. You want planning, you can do monte carlo tr…

There's a recently released kungfu movie whose story happened in the Kowloon walled city. The government is planning to relocate the movie set to the original address for exhibitio…

Maybe it's easier to understand in the format of annotated code https://nlp.seas.harvard.edu/2018/04/03/attention.html

There are way more text training data than voice data. It also allows you to use all the benchmarks and tool integrations that have already been developed for LLMs.

How do I know if it is the original "Where's Waldo" under the paper?

Some "agents" like the minecraft bot Voyager( https://github.com/MineDojo/Voyager ) have a control loop, they are given a high level task and then they use LLM to decide what actio…

I am not sure what you mean by "turn these one-shot APIs into Markov chains." To me, langchain was mostly marketed as a framework that makes RAG easy by providing integration with …

The current gen llms tokenize numbers digit by digit unlike earlier llms.

I recommend reading https://github.com/bkitano/llama-from-scratch over the article op linked. It actually teaches you how to build llama iteratively, test, debug and interpret the …

Write a chrome extension that prompts a LLM/uses a custom trained (BERT based) text classification model to filter the posts. https://github.com/thomasj02/AiFilter https://github.c…