LitBench: A Benchmark and Dataset for Reliable Evaluation of Creative Writing
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918077986242560 |
|---|---|
| author | Fein, Daniel Russo, Sebastian Xiang, Violet Jolly, Kabir Rafailov, Rafael Haber, Nick |
| author_facet | Fein, Daniel Russo, Sebastian Xiang, Violet Jolly, Kabir Rafailov, Rafael Haber, Nick |
| contents | Evaluating creative writing generated by large language models (LLMs) remains challenging because open-ended narratives lack ground truths. Without performant automated evaluation methods, off-the-shelf (OTS) language models are employed as zero-shot judges, yet their reliability is unclear in this context. In pursuit of robust evaluation for creative writing, we introduce LitBench, the first standardized benchmark and paired dataset for creative writing verification, comprising a held-out test set of 2,480 debiased, human-labeled story comparisons drawn from Reddit and a 43,827-pair training corpus of human preference labels. Using LitBench, we (i) benchmark zero-shot LLM judges, (ii) train Bradley Terry and generative reward models, and (iii) conduct an online human study to validate reward model rankings on newly LLM-generated stories. Our benchmark identifies Claude-3.7-Sonnet as the strongest off-the-shelf judge, reaching 73% agreement with human preferences; among trained reward models, Bradley-Terry and Generative reward models both attain an accuracy of 78%, outperforming all off-the-shelf judges. An online human study further confirms that our trained reward models consistently align with human preferences in novel LLM-generated stories. We release LitBench and reward models at https://huggingface.co/collections/SAA-Lab/litbench-68267b5da3aafe58f9e43461, providing a vetted resource for reliable, automated evaluation and optimization of creative writing systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_00769 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LitBench: A Benchmark and Dataset for Reliable Evaluation of Creative Writing Fein, Daniel Russo, Sebastian Xiang, Violet Jolly, Kabir Rafailov, Rafael Haber, Nick Computation and Language Artificial Intelligence Evaluating creative writing generated by large language models (LLMs) remains challenging because open-ended narratives lack ground truths. Without performant automated evaluation methods, off-the-shelf (OTS) language models are employed as zero-shot judges, yet their reliability is unclear in this context. In pursuit of robust evaluation for creative writing, we introduce LitBench, the first standardized benchmark and paired dataset for creative writing verification, comprising a held-out test set of 2,480 debiased, human-labeled story comparisons drawn from Reddit and a 43,827-pair training corpus of human preference labels. Using LitBench, we (i) benchmark zero-shot LLM judges, (ii) train Bradley Terry and generative reward models, and (iii) conduct an online human study to validate reward model rankings on newly LLM-generated stories. Our benchmark identifies Claude-3.7-Sonnet as the strongest off-the-shelf judge, reaching 73% agreement with human preferences; among trained reward models, Bradley-Terry and Generative reward models both attain an accuracy of 78%, outperforming all off-the-shelf judges. An online human study further confirms that our trained reward models consistently align with human preferences in novel LLM-generated stories. We release LitBench and reward models at https://huggingface.co/collections/SAA-Lab/litbench-68267b5da3aafe58f9e43461, providing a vetted resource for reliable, automated evaluation and optimization of creative writing systems. |
| title | LitBench: A Benchmark and Dataset for Reliable Evaluation of Creative Writing |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2507.00769 |