HN user

isaacfung

241 karma
Posts0
Comments105
View on HN
No posts found.
Inductive or deductive? Rethinking the fundamental reasoning abilities of LLMs 2 years ago

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 then the LLM has to answer the original question based on the Chinese translation.

It's a flaw of the tokenization we choose. We can train an LLM using letters instead of tokens as the base units but that would be inefficient.

0ThreadHN
Diffusion models are real-time game engines 2 years ago

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 on your shoulder(with body motion measured by some sesnors though the action space would be much larger).

Such a "game engine" can potentially be used as a simulation gym environment to train RL agents.

0ThreadHN
'Model collapse'? An expert explains the rumours about an impending AI doom 2 years ago

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 means to interact with the environment so they can obtain new training data?

e.g. The minecraft bot Voyager can explore the game environment and extend its skill library(stored as a vector database), is that considered as innovation? There are also systems like leandojo/alphaproof that discover new proofs and use LLM in non-trivial ways(not just naively predict the next token in one shot). Reinforcement learning algorithms like AlphaGo/AlphaZero use self play and use monte carlo tree search to learn to outperform humans. You can similarly use LLMs to generate actions and estimate state values(check the language agent tree search paper).

Most people use LLMs by prompting them with some additional context(chat history and data retrived from database) but there is nothing that stops us from continuously improving a LLM(either by modifying its weight or augmenting it with external database) by asking it to evaluate the task outcome/error message and feeding it back to the LLM. We can also ask it to just keep on generating new tasks to experiment with the environment/internet to get new knowledge.

0ThreadHN
Betting on DSPy for Systems of LLMs 2 years ago

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(according to benchmarks) improve the performance of coding and math related tasks.

0ThreadHN
Overcoming the limits of current LLMs 2 years ago

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 tree search and use LLM to evaluate which node to explore next. You want verifiable reasoning, you can ask it to generate code(an approach used by recent AI olympiad winner and many previous papers).

What is even "planning", finding desirable/optimal solutions to some constrained satisfaction problems? Is the llm based minecraft bot voyager not doing some kind of planning?

LLMs have their limitations. Then augment them with external data sources, code interpreters, give it ways to interact with real world/simulation environment.

0ThreadHN
Why we no longer use LangChain for building our AI agents 2 years ago

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 all kinds of data sources(vector db, pdf, sql db, web search, etc). Also older models(including initial chatgpt) had limited context lengths. Langchain helped you to manage the conversation memory by splitting it up and storing the pieces in a vector db. Another thing langchain did was implementing the react framework(which you can implement with a few lines of code) to help you answer multi hop problems.

0ThreadHN
BASE TTS: The largest text-to-speech model to-date 2 years ago

There are lots of video content with audio. We can train a facial expression classification model to detect the speaker's emotion(we can also use a multimodal model to take in consideration of the language context).

Another potential source of data is voice acting script of animations. I always thought the storyboards of films/animations can be great annotated training data but it seems there are no open datasets, probably because of copyright issues.

0ThreadHN
Antagonistic AI 2 years ago

It seems we can add antagonistic AI to correct many kinds of undesirable human behaviors

Browser extension that monitors how much time I have wasted on social media/porn

Browser extension that reminds me when I made a comment without reading a linked article(like right now)

Imagine a social media platform with an AI that will spot the most stupid comment and shame it using logical arguments and factual evidences, it can even dig up the user comment history to expose their hypocrisy

0ThreadHN