# DeltaScore: Fine-Grained Story Evaluation with Perturbations

arXiv: https://arxiv.org/abs/2303.08991 · HTML: https://mlsanity.com/paper/2303.08991/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 53 references that are in the map.
- First posted on arXiv 15 Mar 2023.

## Builds on (most cited first, 30 of 39)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [BLOOM: A 176B-Parameter Open-Access Multilingual Language Model](https://mlsanity.com/paper/2211.05100/) (2022)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [BARTScore: Evaluating Generated Text as Text Generation](https://mlsanity.com/paper/2106.11520/) (2021)
- [BERT-ATTACK: Adversarial Attack Against BERT Using BERT](https://mlsanity.com/paper/2004.09984/) (2020)
- [MoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distance](https://mlsanity.com/paper/1909.02622/) (2019)
- [GPTScore: Evaluate as You Desire](https://mlsanity.com/paper/2302.04166/) (2023)
- [Towards a Unified Multi-Dimensional Evaluator for Text Generation](https://mlsanity.com/paper/2210.07197/) (2022)
- [Re3: Generating Longer Stories With Recursive Reprompting and Revision](https://mlsanity.com/paper/2210.06774/) (2022)
- [A Survey of Evaluation Metrics Used for NLG Systems](https://mlsanity.com/paper/2008.12009/) (2020)
- [A Knowledge-Enhanced Pretraining Model for Commonsense Story Generation](https://mlsanity.com/paper/2001.05139/) (2020)
- [Automatic Machine Translation Evaluation in Many Languages via Zero-Shot Paraphrasing](https://mlsanity.com/paper/2004.14564/) (2020)
- [The Perils of Using Mechanical Turk to Evaluate Open-Ended Text Generation](https://mlsanity.com/paper/2109.06835/) (2021)
- [Is GPT-3 Text Indistinguishable from Human Text? Scarecrow: A Framework for Scrutinizing Machine Text](https://mlsanity.com/paper/2107.01294/) (2021)
- [Of Human Criteria and Automatic Metrics: A Benchmark of the Evaluation of Story Generation](https://mlsanity.com/paper/2208.11646/) (2022)
- [MEGATRON-CNTRL: Controllable Story Generation with External Knowledge Using Large-Scale Language Models](https://mlsanity.com/paper/2010.00840/) (2020)
- [OpenMEVA: A Benchmark for Evaluating Open-ended Story Generation Metrics](https://mlsanity.com/paper/2105.08920/) (2021)
- [UNION: An Unreferenced Metric for Evaluating Open-ended Story Generation](https://mlsanity.com/paper/2009.07602/) (2020)
- [Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation](https://mlsanity.com/paper/2109.06379/) (2021)
- [The Next Chapter: A Study of Large Language Models in Storytelling](https://mlsanity.com/paper/2301.09790/) (2023)
- [Perturbation CheckLists for Evaluating NLG Evaluation Metrics](https://mlsanity.com/paper/2109.05771/) (2021)
- [On the Blind Spots of Model-Based Evaluation Metrics for Text Generation](https://mlsanity.com/paper/2212.10020/) (2022)

## Built on by (most cited first, 9 of 9)

- [LLM-based NLG Evaluation: Current Status and Challenges](https://mlsanity.com/paper/2402.01383/) (2024)
- [FIRE: Fact-checking with Iterative Retrieval and Verification](https://mlsanity.com/paper/2411.00784/) (2024)
- [Are LLM-based Evaluators Confusing NLG Quality Criteria?](https://mlsanity.com/paper/2402.12055/) (2024)
- [What Makes a Good Story and How Can We Measure It? A Comprehensive Survey of Story Evaluation](https://mlsanity.com/paper/2408.14622/) (2024)
- [Creativity in AI: Progresses and Challenges](https://mlsanity.com/paper/2410.17218/) (2024)
- [What Matters in Evaluating Book-Length Stories? A Systematic Study of Long Story Evaluation](https://mlsanity.com/paper/2512.12839/) (2025)
- [EvolvR: Self-Evolving Pairwise Reasoning for Story Evaluation to Enhance Generation](https://mlsanity.com/paper/2508.06046/) (2025)
- [LecEval: An Automated Metric for Multimodal Knowledge Acquisition in Multimedia Learning](https://mlsanity.com/paper/2505.02078/) (2025)
- [RusFinChain: A Russian Benchmark for Verifiable Chain-of-Thought Reasoning in Finance with Fuzzy-Aligned Evaluation](https://mlsanity.com/paper/2607.01388/) (2026)

## Abstract (arXiv)

Numerous evaluation metrics have been developed for natural language generation tasks, but their effectiveness in evaluating stories is limited as they are not specifically tailored to assess intricate aspects of storytelling, such as fluency and interestingness. In this paper, we introduce DELTASCORE, a novel methodology that employs perturbation techniques for the evaluation of nuanced story aspects. Our central proposition posits that the extent to which a story excels in a specific aspect (e.g., fluency) correlates with the magnitude of its susceptibility to particular perturbations (e.g., the introduction of typos). Given this, we measure the quality of an aspect by calculating the likelihood difference between pre- and post-perturbation states using pre-trained language models. We compare DELTASCORE with existing metrics on storytelling datasets from two domains in five fine-grained story aspects: fluency, coherence, relatedness, logicality, and interestingness. DELTASCORE demonstrates remarkable performance, revealing a surprising finding that a specific perturbation proves highly effective in capturing multiple aspects.