datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
story_generation_sft
Story Generation SFT (EpisodeBench)
This dataset is the supervised fine-tuning (SFT) training resource released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
EpisodeBench represents each story as an episode graph with explicit states, observable trigger-conditioned transitions, and interaction budgets, turning long-form narrative progression into a measurable evaluation object. The Story Generation… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_sft.task105_story_cloze-rocstories_sentence_generation
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task105_story_cloze-rocstories_sentence_generation
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task105_story_cloze-rocstories_sentence_generation.task269_csrg_counterfactual_story_generation
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task269_csrg_counterfactual_story_generation
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task269_csrg_counterfactual_story_generation.RUCAIBox-Story-Generation-Alpacahttps://huggingface.co/datasets/RUCAIBox/Story-Generation
RUC AI Box HC Story Generation augmented and converted to alpaca format.
No filtering has been done.
story-generation
Story generation
Dataset Summary
This dataset contains summaries and stories from RUCAIBox/Story-Generation dataset.
Dataset Structure
Data Fields
summary: The summary of the story
story: The story
story_generation_reward_train_exppos
Reward Training — Exppos (EpisodeBench)
This dataset is one of four distribution-controlled reward-training resources released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
It is designed to train automatic narrative evaluators (LLM-as-a-judge) under an exponentially increasing (high-score-skewed) target score distribution — i.e., score frequencies grow with rubric score, so high-quality bands are more… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_reward_train_exppos.story_generation_reward_train_normal
Reward Training — Normal (EpisodeBench)
This dataset is one of four distribution-controlled reward-training resources released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
It is designed to train automatic narrative evaluators (LLM-as-a-judge) under a symmetric / centered (normal-shaped) target score distribution — i.e., score frequencies are concentrated around the rubric mid-point and decay smoothly… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_reward_train_normal.story_generation_rl
Story Generation RL (EpisodeBench)
This dataset is the reinforcement-learning (RL) training resource released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
EpisodeBench represents each story as an episode graph with explicit states, observable trigger-conditioned transitions, and interaction budgets, turning long-form narrative progression into a measurable evaluation object. The Story Generation RL… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_rl.creative_story_generation_dataset
Creative Story Generation Dataset
This dataset contains the five-sentence human and AI creative stories and their expert/non-expert ratings across multiple dimensions from the Evaluating Creative Short Story Generation in Humans and LLMs.
Citation
@misc{ismayilzada2024evaluatingcreativeshortstory,
title={Evaluating Creative Short Story Generation in Humans and Large Language Models},
author={Mete Ismayilzada and Claire Stevenson and Lonneke van der Plas}… See the full description on the dataset page: https://huggingface.co/datasets/mismayil/creative_story_generation_dataset.story_generation_reward_test
Reward Test — Held-Out Evaluation Set (EpisodeBench)
This dataset is the held-out test set for automatic narrative evaluators released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
It is designed to measure how well an LLM-as-a-judge calibrates to EpisodeBench's synthesized rubric targets. Specifically, the paper reports the average absolute gap between each evaluator's predicted score and the synthesized… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_reward_test.story_generation_reward_train_expneg
Reward Training — Expneg (EpisodeBench)
This dataset is one of four distribution-controlled reward-training resources released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
It is designed to train automatic narrative evaluators (LLM-as-a-judge) under an exponentially decreasing (low-score-skewed) target score distribution — i.e., score frequencies decay with rubric score, so low-quality bands are more… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_reward_train_expneg.story_generation_reward_train_uniform
Reward Training — Uniform (EpisodeBench)
This dataset is one of four distribution-controlled reward-training resources released as part of EpisodeBench, a full-cycle benchmarking pipeline for long-form interactive story generation with controllable RL.
It is designed to train automatic narrative evaluators (LLM-as-a-judge) under a uniform target score distribution — i.e., score frequencies are flattened across the rubric scale, so that low / mid / high quality bands are roughly… See the full description on the dataset page: https://huggingface.co/datasets/HeAAAAA/story_generation_reward_train_uniform.Story-Generation
🇰🇿 Stories and Dialogue Generation
📖 Overview
Stories Generation is a creative writing dataset specifically curated for the Kazakh language.
📊 Dataset Statistics
General Metrics
Metric
Count
Total Samples
400
Total Words (approx.)
109,735
Avg. Words per Sample
274
Word Count Distribution (Per Field)
The following table details the distribution of word counts across different fields in the… See the full description on the dataset page: https://huggingface.co/datasets/farabi-lab/Story-Generation.storyGeneration
Dataset Card for "storyGeneration"
More Information needed
RUCAIBox-Story-Generation-teststory-generation-datasetqwen3_0.6b-rlvr_task269_csrg_counterfactual_story_generationtask059_ropes_story_generation
Dataset Card for Natural Instructions (https://github.com/allenai/natural-instructions) Task: task059_ropes_story_generation
Additional Information
Citation Information
The following paper introduces the corpus in detail. If you use the corpus in published work, please cite it:
@misc{wang2022supernaturalinstructionsgeneralizationdeclarativeinstructions,
title={Super-NaturalInstructions: Generalization via Declarative Instructions on 1600+ NLP Tasks}… See the full description on the dataset page: https://huggingface.co/datasets/Lots-of-LoRAs/task059_ropes_story_generation.Horror_Story_Generationqwen3_0.6b-rlvr_task105_story_cloze-rocstories_sentence_generationflan_combined_task105_story_cloze-rocstories_sentence_generation