# Evaluation Under Imperfect Benchmarks and Ratings: A Case Study in Text Simplification

arXiv: https://arxiv.org/abs/2504.09394 · HTML: https://mlsanity.com/paper/2504.09394/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 50 references that are in the map.
- First posted on arXiv 13 Apr 2025.

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

- [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)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [BERTScore: Evaluating Text Generation with BERT](https://mlsanity.com/paper/1904.09675/) (2019)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate](https://mlsanity.com/paper/2308.07201/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [Deep Learning Scaling is Predictable, Empirically](https://mlsanity.com/paper/1712.00409/) (2017)
- [MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers](https://mlsanity.com/paper/2102.01454/) (2021)
- [Mixture-of-Agents Enhances Large Language Model Capabilities](https://mlsanity.com/paper/2406.04692/) (2024)
- [Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models](https://mlsanity.com/paper/2404.18796/) (2024)
- [Inverse Scaling: When Bigger Isn't Better](https://mlsanity.com/paper/2306.09479/) (2023)
- [Attributed Question Answering: Evaluation and Modeling for Attributed Large Language Models](https://mlsanity.com/paper/2212.08037/) (2022)
- [Sentence Simplification with Deep Reinforcement Learning](https://mlsanity.com/paper/1703.10931/) (2017)
- [Neural CRF Model for Sentence Alignment in Text Simplification](https://mlsanity.com/paper/2005.02324/) (2020)
- [ASSET: A Dataset for Tuning and Evaluation of Sentence Simplification Models with Multiple Rewriting Transformations](https://mlsanity.com/paper/2005.00481/) (2020)
- [An Empirical Study of LLM-as-a-Judge for LLM Evaluation: Fine-tuned Judge Model is not a General Substitute for GPT-4](https://mlsanity.com/paper/2403.02839/) (2024)
- [BLEU is Not Suitable for the Evaluation of Text Simplification](https://mlsanity.com/paper/1810.05995/) (2018)
- [EditNTS: An Neural Programmer-Interpreter Model for Sentence Simplification through Explicit Editing](https://mlsanity.com/paper/1906.08104/) (2019)
- [Do LLMs write like humans? Variation in grammatical and rhetorical styles](https://mlsanity.com/paper/2410.16107/) (2024)
- [LENS: A Learnable Evaluation Metric for Text Simplification](https://mlsanity.com/paper/2212.09739/) (2022)
- [Controllable Text Simplification with Explicit Paraphrasing](https://mlsanity.com/paper/2010.11004/) (2020)
- [Evaluating Factuality in Text Simplification](https://mlsanity.com/paper/2204.07562/) (2022)
- [Is GPT-4 a reliable rater? Evaluating Consistency in GPT-4 Text Ratings](https://mlsanity.com/paper/2308.02575/) (2023)
- [Simplicity Level Estimate (SLE): A Learned Reference-Less Metric for Sentence Simplification](https://mlsanity.com/paper/2310.08170/) (2023)
- [Inverse Visual Question Answering: A New Benchmark and VQA Diagnosis Tool](https://mlsanity.com/paper/1803.06936/) (2018)
- [REFeREE: A REference-FREE Model-Based Metric for Text Simplification](https://mlsanity.com/paper/2403.17640/) (2024)

## Built on by (most cited first, 4 of 4)

- [AI Evaluation Should Require Standardized Item-Level Data Releases](https://mlsanity.com/paper/2604.03244/) (2026)
- [German4All -- A Dataset and Model for Readability-Controlled Paraphrasing in German](https://mlsanity.com/paper/2508.17973/) (2025)
- [Policy-based Sentence Simplification: Replacing Parallel Corpora with LLM-as-a-Judge](https://mlsanity.com/paper/2512.06228/) (2025)
- [Profiling German Text Simplification with Interpretable Model-Fingerprints](https://mlsanity.com/paper/2601.13050/) (2026)

## Abstract (arXiv)

Despite the successes of language models, their evaluation remains a daunting challenge for new and existing tasks. We consider the task of text simplification, commonly used to improve information accessibility, where evaluation faces two major challenges. First, the data in existing benchmarks might not reflect the capabilities of current language models on the task, often containing disfluent, incoherent, or simplistic examples. Second, existing human ratings associated with the benchmarks often contain a high degree of disagreement, resulting in inconsistent ratings; nevertheless, existing metrics still have to show higher correlations with these imperfect ratings. As a result, evaluation for the task is not reliable and does not reflect expected trends (e.g., more powerful models being assigned higher scores). We address these challenges for the task of text simplification through three contributions. First, we introduce SynthSimpliEval, a synthetic benchmark for text simplification featuring simplified sentences generated by models of varying sizes. Through a pilot study, we show that human ratings on our benchmark exhibit high inter-annotator agreement and reflect the expected trend: larger models produce higher-quality simplifications. Second, we show that auto-evaluation with a panel of LLM judges (LLMs-as-a-jury) often suffices to obtain consistent ratings for the evaluation of text simplification. Third, we demonstrate that existing learnable metrics for text simplification benefit from training on our LLMs-as-a-jury-rated synthetic data, closing the gap with pure LLMs-as-a-jury for evaluation. Overall, through our case study on text simplification, we show that a reliable evaluation requires higher quality test data, which could be obtained through synthetic data and LLMs-as-a-jury ratings.