8 ms·
It's always fun to look at the prompts used by these projects. Here are a few snippets from this one: >You are a helpful assistant that tells me the next immed
by Imnimo 3y ago
It's always fun to look at the prompts used by these projects. Here are a few snippets from this one:
>You are a helpful assistant that tells me the next immediate task to do in Minecraft. My ultimate goal is to discover as many diverse things as possible, accomplish as many diverse tasks as possible and become the best Minecraft player in the world.
>8) Tasks that require information beyond the player's status to verify should be avoided. For instance, "Placing 4 torches" and "Dig a 2x1x2 hole" are not ideal since they require visual confirmation from the screen. All the placing, building, planting, and trading tasks should be avoided. Do not propose task starting with these keywords
>7) Use `exploreUntil(bot, direction, maxDistance, callback)` when you cannot find something. You should frequently call this before mining blocks or killing mobs. You should select a direction at random every time instead of constantly using (1, 0, 1).
>9) Do not write infinite loops or recursive functions.
You can really imagine the sorts of pitfalls the agent fell into that induced the authors to add these stipulations.
- qayxc 3y agoIn the end it all looks like classical programming with extra steps to me :)
- Imnimo 3y agoEven more than the complex prompts, they give the bot access to a variety of functions implementing primitive actions, which are crafted to give helpful hints if the bot is struggling. For example, the code for mining a block: https://github.com/MineDojo/Voyager/blob/main/voyager/control_primitives/mineBlock.js https://github.com/MineDojo/Voyager/blob/main/voyager/contro... This function keeps a global count of how many times it's been called to mine a block that doesn't exist nearby, and will warn the bot that it should try exploring first instead. There's nothing necessarily wrong with all that - it's an important research question to understand how much hand-holding the agent needs to be able to do these sorts of tasks. But readers should be aware that it's hardly dropping an agent into the world and leaving it to its own devices.
- sterlind 3y agoI do love their use of `bot.chat()` as, like, throwing an exception to the AI. maybe this finally gives developers an incentive to document their code and throw intelligible errors - the machines need it to figure out what they're doing wrong!
- ineedasername 3y agoChatGPT: I am sorry for the confusion. I am a large language model (LLM) trained on natural language and my training data cutoff is September 2021. Hexadecimal references to memory addresses are not part of my training data set.”
- uw_rob 3y agoSome would call this Software 2.0 https://karpathy.medium.com/software-2-0-a64152b37c35 https://karpathy.medium.com/software-2-0-a64152b37c35
- doctoboggan 3y ago>9) Do not write infinite loops or recursive functions. did GPT-4 just solve the halting problem?
- geoelectric 3y agoSeems slightly easier for me to guarantee something I wrote halts. The halting problem is more about a process analyzing someone else's arbitrary code.
- narrator 3y agoIf the brain is like a computer, how do humans even solve the halting problem? Maybe it's because humans feel anxiety at the unproductive passing of time while computers will just do what they're doing forever.
- i2cmaster 3y agoHumans tend to communicate/write software in what are really sub languages. Often these sub languages are not recursive.
- IIAOPSW 3y agoNothing you think about consciously has higher thread priority than the stay-alive loop.
- ethanbond 3y agoThey don’t really. People have different thresholds for acceptable exploration costs and heuristics for approximating those costs (as well as the opportunity cost of alternative courses of action). IMO, in biological systems the explore vs exploit tradeoff is pretty analogous to the halting problem. There doesn’t seem to be an optimal general “solution” to it. Once an organism is very familiar with its environment it’ll approximate near-optimal trade offs, which would suggest it’s just using heuristics.
- tmountain 3y agoHow did you get the prompt?
- cjbprime 3y agohttps://github.com/MineDojo/Voyager/blob/main/voyager/prompts/action_template.txt https://github.com/MineDojo/Voyager/blob/main/voyager/prompt...
- kodah 3y agoIt would be really cool if they simplified an NPCs world to include GPT prompts for an RPG-like experience.
- 1270018080 3y agoThat's the only application I see for black box language models that hallucinate text.
- neuronexmachina 3y agoSomething like this? https://arstechnica.com/information-technology/2023/04/surprising-things-happen-when-you-put-25-ai-agents-together-in-an-rpg-town/ https://arstechnica.com/information-technology/2023/04/surpr...