Sign in to view source links and access this dataset
Description
A benchmark dataset for evaluating novel-to-film generation, created by ZuoHaotong and last updated on July 21, 2026. It contains 15 source novels grouped into three difficulty levels—easy, medium, and hard—with five novels per level. The dataset includes generated film or video outputs from multiple inference systems for comparison.
Use Cases
Benchmarking text-to-video generation systems based on the provided source novels.
Comparing the performance of different inference models on transforming written stories into short films.
Evaluating generative AI outputs across varying levels of narrative complexity, as defined by the easy, medium, and hard difficulty tiers.
Strengths
Contains 15 distinct source novels providing a foundation for evaluation.
Structures evaluation by three explicit difficulty levels: easy, medium, and hard.
Includes outputs from multiple inference systems, enabling comparative analysis.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count is unknown, which may limit suitability assessment.
Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
Source
ZuoHaotong on Hugging Face
Collection Method
Likely compiled for AI benchmarking purposes; specific gathering method unknown.
Freshness
Last updated 2026-07-21 14:21:41; freshness should be verified.
License is unknown; users should verify usage rights before download.