Long Text Generation Challenge
This addresses the problem of generating coherent long texts for researchers in natural language processing, though it is incremental as it builds on existing text generation methods.
The paper introduces the LTG Challenge, a shared task for generating consistent, human-like long texts (specifically Harry Potter fanfiction) from 1000-token prompts, and proposes a novel statistical metric called GAPELMAPER and a human evaluation protocol to assess text structuredness.
We propose a shared task of human-like long text generation, LTG Challenge, that asks models to output a consistent human-like long text (a Harry Potter generic audience fanfic in English), given a prompt of about 1000 tokens. We suggest a novel statistical metric of the text structuredness, GloVe Autocorrelations Power/ Exponential Law Mean Absolute Percentage Error Ratio (GAPELMAPER) and a human evaluation protocol. We hope that LTG can open new avenues for researchers to investigate sampling approaches, prompting strategies, autoregressive and non-autoregressive text generation architectures and break the barrier to generate consistent long (40K+ token) texts.