They have a fairly simple interface compared with traditional apps, as they only require users to chat, and the chatbots are supposed to understand and do whatever the user demands from them, at least in theory. It also couldn’t have been possible without OpenAI and their released GPT-2 models.Chatbots are often touted as a revolution in the way users interact with technology and businesses. Note: AI Dungeon 1 and 2 were created by Nick Walton with support from David Wingate, Max Robinson, and Alan Walton. You can read more stories from AI Dungeon 2 and play the game yourself here. You thank him again before leaving.When you get outside you find yourself staring up at a massive stone wall.Īlthough AI Dungeon 2 has these limitations it is still a remarkable demonstration of language models’ ability to generate stories and a dramatic advancement in AI generated interactive fiction. With those words the king begins to leave. “Thank you for the equipment, I will certainly look into this”. He tells you that he has plenty of gold in his treasury so you should just ask him for it. You nod and explain that you need some new equipment. Other times the model has difficulty keeping track of who is who, especially in dialogue. You sniff the tree and find that its scent is very pleasant. You immediately feel a strong sense of euphoria. You take a deep breath and stare intently at the tree. It seems to be made of wood, but its leaves seem to be green rather than normal leafy green. You walk up to it and examine it closely. You find it completely deserted save for a single tree. The model at times shows remarkable understanding and writing quality like in this sample where the player tries to find love on a deserted island. We found N=8 to be a good amount of memory to feed the model. He replies.įor each action you type the model is fed the context sentence as well as the past N action-result pairs in its memory to generate the result. “My name is Trelik, but you may call me whatever you wish”. It looks like an old man with white hair and beard. However, you know it’s not human because of the strange eyes staring out from its face. It’s very old and has many cracks in its facade. The voice comes from behind a large stone pillar. “Hm, well maybe you can find something there”. “I guess I should go back to my home now”. You finish your long journey and finally arrive at the ruin you’ve been looking for. You are Zoltan, a wizard living in the kingdom of Larion. For example if you choose to be a wizard named Zoltan in a fantasy world your adventure might begin like this. The context sentence is always included as input to the model to help keep it grounded to the chosen setting while the starting prompt is the beginning of the story.
These choices generate a starting prompt and context sentence. When playing AI Dungeon 2 the player can choose from several possible settings and characters.
#Python free text adventure download upgrade#
We also upgrade the size of our model to OpenAI’s largest 1.5B parameter model and fine-tune it on a collection of text adventures obtained from .įollowing the example of the Salesforce CTRL model we also modified the sampling of our model to add a penalty to already generated words to reduce the repetition issues that GPT-2 has and enable a lower temperature setting. The model then continues generating the story resulting from that action. In AI Dungeon 2 we do away with pregenerated actions and allow the user to enter any action. In the interest of computational cost, possible actions and their results were generated and given to the player to choose from. In the first AI Dungeon, we created and deployed a deep learning generated text adventure using OpenAI’s 124M parameter GPT-2 model. What if you could also choose any action you can think of instead of being limited by the imagination of the developers who created the game? Imagine an infinitely generated world that you could explore endlessly, continually finding entirely new content and adventures.
0 kommentar(er)
