Hacker News: levesque

New comment by levesque in "Apple M4 benchmarks suggest it is the new single-core performance champ"

levesque — Fri, 10 May 2024 12:49:01 +0000

Is there really no way to do this?

New comment by levesque in "‘No way out’: how video games use tricks from gambling to attract big spenders"

levesque — Thu, 20 Jul 2023 10:50:16 +0000

To be fair, the people quoted in the book don't seem to be having that much fun. But yes, most of modern tech seems to be exploiting the same mechanism, generating addiction.

New comment by levesque in "Game design wiki"

levesque — Fri, 14 Jul 2023 18:25:58 +0000

Absolutely, was thinking the same. IIRC it's also only telling success stories, so people burning themselves out but succeeding in the end.

New comment by levesque in "As an AI Language Model"

levesque — Tue, 25 Apr 2023 13:09:59 +0000

For HN and forums in general, I think this will mean disabling APIs and having strict captchas for posting.

Beyond HN, I think this will translate in video content and reviews becoming more trustworthy, even if it's just a person reading a LLM-produced script. You will at least know they cared enough to put a human in the loop. That and reputation. More and more credit will be assigned based on reputation, number of followers, etc. And that'll be until each of these systems get cracked somehow (fake followers, plausible generated videos, etc.).

New comment by levesque in "As an AI Language Model"

levesque — Mon, 24 Apr 2023 13:05:14 +0000

Predicting whether a text was written by a LLM or not is not trivial. What was the latest number by OpenAI? 30%? As LLMs get better, it seems like we won't be able to distinguish real text from fake text. Your LLM will be able to summarize it, but it will still be 99% spam.

New comment by levesque in "StableLM: A new open-source language model"

levesque — Wed, 19 Apr 2023 19:15:44 +0000

Ah, interesting. Thought GPT-3.5 had the same structure as GPT-3, for some reason. GPT-4 would obviously be different.

New comment by levesque in "StableLM: A new open-source language model"

levesque — Wed, 19 Apr 2023 17:37:36 +0000

Isn't ChatGPT a 165B parameter model?

New comment by levesque in "HuggingGPT: Solving AI tasks with ChatGPT and its friends in HuggingFace"

levesque — Fri, 31 Mar 2023 18:09:18 +0000

This read more like a technical demo than a scientific paper. Wonder why they put it on arXiv.

New comment by levesque in "AI-enhanced development makes me more ambitious with my projects"

levesque — Fri, 31 Mar 2023 12:21:44 +0000

It's not only about performance, it's also about maintainability. Codebases written by inexperienced programmers are extremely convoluted, nothing is decoupled and functions are extremely long and do multiple things. Extremely hard code to maintain or add features to.

New comment by levesque in "Ok, it’s time to freak out about AI"

levesque — Thu, 16 Mar 2023 18:15:19 +0000

Anthropomorphization of a text completion engine. Humanity will not be destroyed by a fancy autocomplete bot. This is just alarmist clickbait, moving on.

New comment by levesque in "Open source implementation for LLaMA-based ChatGPT"

levesque — Mon, 27 Feb 2023 16:30:09 +0000

Yes and no. You wouldn't say GPT to mean large language models or autoregressive language models. I would've thought the same to be true for ChatGPT instead of Chatbots with RL from human feedback (RLHF), perhaps the field is moving towards adopting ChatGPT as a paradigm name. Note that the title doesn't say a ChatGPT-like model based on LLaMa, it says outright opensource implementation of ChatGPT.

New comment by levesque in "Open source implementation for LLaMA-based ChatGPT"

levesque — Mon, 27 Feb 2023 15:57:09 +0000

I think I prefer the other commenter's point, referring to ChatGPT as a known learning paradigm for chatbots. But thanks for the little crash course on analogies ;)

New comment by levesque in "Open source implementation for LLaMA-based ChatGPT"

levesque — Mon, 27 Feb 2023 14:53:01 +0000

It uses a different engine, so this is as related to ChatGPT as a Toyota Corolla is related to a BMW car. This is an efficient and open-source chatbot, which is very good news, but the authors just wrote a clickbait title and they know it.

New comment by levesque in "Open source implementation for LLaMA-based ChatGPT"

levesque — Mon, 27 Feb 2023 14:45:02 +0000

In what way is this a ChatGPT implementation or equivalent? Seems like a chatbot based on a different backend, therefore it has absolutely zero link to ChatGPT.

New comment by levesque in "MarioGPT Uses AI to Generate Endless Super Mario Levels for Free"

levesque — Wed, 15 Feb 2023 15:55:00 +0000

AI models might be able to do this in the future by themselves, though with current paradigms AI will barely generate copies of existing levels with little creativity. Sure, a composition of existing level pieces could lead to an interesting level design, though it would be more by accident than by design. Models do not maximize player enjoyment, there is no metric for that. Maybe engagement metrics could be used, but I don't think players would stick around long enough playing bad levels to reach a viable model.

New models and paradigms will come up, but until then I'd say anything AI-generated will feel pretty vanilla and somewhat incoherent.

New comment by levesque in "MarioGPT Uses AI to Generate Endless Super Mario Levels for Free"

levesque — Wed, 15 Feb 2023 15:18:54 +0000

Yeah OP missed the target on this one. Super Mario Maker is a great example of insane creativity demonstrated by a community. AI can't do that though.

New comment by levesque in "Jupyterlab Desktop"

levesque — Sun, 12 Feb 2023 14:33:33 +0000

Notebooks are complex because they allow you to mix different cell types and keep outputs. That's not a notebook.

New comment by levesque in "AI is coming to Hollywood with Metaphysic"

levesque — Thu, 02 Feb 2023 16:49:39 +0000

What an obnoxious title and article. AI has been part of the Hollywood pipeline for a while, they have always been using bleeding edge technology, sometimes AI-assisted.

New comment by levesque in "What if AI didn't make you a bad writer, but a better thinker?"

levesque — Thu, 26 Jan 2023 14:12:05 +0000

It wasn't so much a critic of your article as thinking about coping mechanisms when the volume of articles being put out is significantly increased and nobody will be able or want to deal with all this noise. Fight GPT by GPT :)

New comment by levesque in "What if AI didn't make you a bad writer, but a better thinker?"

levesque — Thu, 26 Jan 2023 13:48:33 +0000

Good news is I will be able to ask GPT-like models to summarize me all these articles and I won't have to sift through them.