Awesome

RecurrentGPT

Framework Illustration

RecurrentGPT replaces the vectorized elements (i.e., cell state, hidden state, input, and output) in a Long-short Term Memory RNN (LSTM) with natural language (i.e., paragraphs of texts), and simulates the recurrence mechanism with prompt engineering.

At each timestep t, RecurrentGPT receives a paragraph of text and a brief plan of the next paragraph, which are both generated in step t − 1. It then attends to the long-term memory, which contains the summaries of all previously generated paragraphs and can be stored on hard drives, and relevant paragraphs can be retrieved with semantic search.

RecurrentGPT also maintains a short-term memory that summarizes key information within recent timesteps in natural language and is updated at each time step. RecurrentGPT combines all aforementioned inputs in a prompt and asks the backbone LLM to generate a new paragraph, a short plan for the next paragraph, and updates the long-short term memory by rewriting the short-term memory and appending the summary of the output paragraph to the long-term memory.

Example

Deployment

You can change the configurations given in the recurrent.sh script

iteration: 10                       #(int) the number of rounds you would like it to roll.
outfile: response.txt               #(str) the output file path.
init_prompt: init_prompt.json       #(str) the path to the prompt used for initialization.
topic: Aliens                       #(str) the topic that you wish your novel is about.
type: science-fiction               #(str) the type of novel you would like to write.

Then after specify your OPENAI_API_KEY in the recurrent.sh file, you can run

sh recurrent.sh

NOTE: If your local internet is not allowed to access OpenAI's API, you might need to first export your HTTP proxy in the recurrent.sh file as well.

export http_proxy='your_proxy'

Showcases

Prompt Engineering

Iterations

Human writer starts by choosing the topic he/she wants to write and writes a short paragraph describing the background and the outline of the book. Then RECURRENTGPT automatically generates the first paragraphs and provides a few possible options for the writer to continue the story. The writer may select one from them and edit it if needed. He or she can also write a short plan for the next few paragraphs by him/herself if generated plans are all inappropriate, which makes human-AI co-writing process more flexible

Web demo

You can directly use our online demo at: https://www.aiwaves.org/recurrentgpt and https://www.aiwaves.org/interactivefiction

Or you can run it on your local machine by editing the OPENAI_API_KEY and OPENAI_Proxy in utils.py and then run:

python gradio_server.py

Use customized LLMs for local deployment

Please refer to https://github.com/jackaduma/Recurrent-LLM to use opensource LLMs for local deployment. Many thanks to @jackaduma

Citation

@misc{zhou2023recurrentgpt,
      title={RecurrentGPT: Interactive Generation of (Arbitrarily) Long Text}, 
      author={Wangchunshu Zhou and Yuchen Eleanor Jiang and Peng Cui and Tiannan Wang and Zhenxin Xiao and Yifan Hou and Ryan Cotterell and Mrinmaya Sachan},
      year={2023},
      eprint={2305.13304},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}