# On the Blind Spots of Model-Based Evaluation Metrics for Text Generation

arXiv: https://arxiv.org/abs/2212.10020 · HTML: https://mlsanity.com/paper/2212.10020/

## Key facts

- Cited by 49 papers in the map (data through 14 Oct 2026).
- Builds on 69 of its 89 references that are in the map.
- First posted on arXiv 20 Dec 2022.

## Builds on (most cited first, 30 of 69)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [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)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [BLEURT: Learning Robust Metrics for Text Generation](https://mlsanity.com/paper/2004.04696/) (2020)
- [Hierarchical Neural Story Generation](https://mlsanity.com/paper/1805.04833/) (2018)
- [Teaching Machines to Read and Comprehend](https://mlsanity.com/paper/1506.03340/) (2015)
- [COMET: A Neural Framework for MT Evaluation](https://mlsanity.com/paper/2009.09025/) (2020)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [BARTScore: Evaluating Generated Text as Text Generation](https://mlsanity.com/paper/2106.11520/) (2021)
- [Cross-Task Generalization via Natural Language Crowdsourcing Instructions](https://mlsanity.com/paper/2104.08773/) (2021)
- [How multilingual is Multilingual BERT?](https://mlsanity.com/paper/1906.01502/) (2019)
- [Evaluating the Factual Consistency of Abstractive Text Summarization](https://mlsanity.com/paper/1910.12840/) (2019)
- [Neural Text Generation with Unlikelihood Training](https://mlsanity.com/paper/1908.04319/) (2019)
- [BERT-ATTACK: Adversarial Attack Against BERT Using BERT](https://mlsanity.com/paper/2004.09984/) (2020)
- [MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontiers](https://mlsanity.com/paper/2102.01454/) (2021)
- [DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts](https://mlsanity.com/paper/2105.03023/) (2021)
- [MoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distance](https://mlsanity.com/paper/1909.02622/) (2019)
- [Asking and Answering Questions to Evaluate the Factual Consistency of Summaries](https://mlsanity.com/paper/2004.04228/) (2020)
- [FUDGE: Controlled Text Generation With Future Discriminators](https://mlsanity.com/paper/2104.05218/) (2021)
- [What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models](https://mlsanity.com/paper/1907.13528/) (2019)

## Built on by (most cited first, 30 of 49)

- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [MT-Bench-101: A Fine-Grained Benchmark for Evaluating Large Language Models in Multi-Turn Dialogues](https://mlsanity.com/paper/2402.14762/) (2024)
- [LLMs as Narcissistic Evaluators: When Ego Inflates Evaluation Scores](https://mlsanity.com/paper/2311.09766/) (2023)
- [LLM-based NLG Evaluation: Current Status and Challenges](https://mlsanity.com/paper/2402.01383/) (2024)
- [CritiqueLLM: Towards an Informative Critique Generation Model for Evaluation of Large Language Model Generation](https://mlsanity.com/paper/2311.18702/) (2023)
- [Uncertainty in Natural Language Generation: From Theory to Applications](https://mlsanity.com/paper/2307.15703/) (2023)
- [Understanding In-Context Learning via Supportive Pretraining Data](https://mlsanity.com/paper/2306.15091/) (2023)
- [Are LLM-based Evaluators Confusing NLG Quality Criteria?](https://mlsanity.com/paper/2402.12055/) (2024)
- [BLESS: Benchmarking Large Language Models on Sentence Simplification](https://mlsanity.com/paper/2310.15773/) (2023)
- [Finding Blind Spots in Evaluator LLMs with Interpretable Checklists](https://mlsanity.com/paper/2406.13439/) (2024)
- [Do GPTs Produce Less Literal Translations?](https://mlsanity.com/paper/2305.16806/) (2023)
- [Stumbling Blocks: Stress Testing the Robustness of Machine-Generated Text Detectors Under Attacks](https://mlsanity.com/paper/2402.11638/) (2024)
- [SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses](https://mlsanity.com/paper/2404.04298/) (2024)
- [The Eval4NLP 2023 Shared Task on Prompting Large Language Models as Explainable Metrics](https://mlsanity.com/paper/2310.19792/) (2023)
- [Automatic Metrics in Natural Language Generation: A Survey of Current Evaluation Practices](https://mlsanity.com/paper/2408.09169/) (2024)
- [Text Generation: A Systematic Literature Review of Tasks, Evaluation, and Challenges](https://mlsanity.com/paper/2405.15604/) (2024)
- [Evaluating LLM Metrics Through Real-World Capabilities](https://mlsanity.com/paper/2505.08253/) (2025)
- [What Makes a Good Story and How Can We Measure It? A Comprehensive Survey of Story Evaluation](https://mlsanity.com/paper/2408.14622/) (2024)
- [APPLS: Evaluating Evaluation Metrics for Plain Language Summarization](https://mlsanity.com/paper/2305.14341/) (2023)
- [Towards Explainable Evaluation Metrics for Machine Translation](https://mlsanity.com/paper/2306.13041/) (2023)
- [ALLURE: Auditing and Improving LLM-based Evaluation of Text using Iterative In-Context-Learning](https://mlsanity.com/paper/2309.13701/) (2023)
- [Diffusion Guided Language Modeling](https://mlsanity.com/paper/2408.04220/) (2024)
- [Verifiable by Design: Aligning Language Models to Quote from Pre-Training Data](https://mlsanity.com/paper/2404.03862/) (2024)
- [DeltaScore: Fine-Grained Story Evaluation with Perturbations](https://mlsanity.com/paper/2303.08991/) (2023)
- [Towards Better Open-Ended Text Generation: A Multicriteria Evaluation Framework](https://mlsanity.com/paper/2410.18653/) (2024)
- [SynthTextEval: Synthetic Text Data Generation and Evaluation for High-Stakes Domains](https://mlsanity.com/paper/2507.07229/) (2025)
- [Is Summary Useful or Not? An Extrinsic Human Evaluation of Text Summaries on Downstream Tasks](https://mlsanity.com/paper/2305.15044/) (2023)
- [IberBench: LLM Evaluation on Iberian Languages](https://mlsanity.com/paper/2504.16921/) (2025)
- [An In-depth Evaluation of Large Language Models in Sentence Simplification with Error-based Human Assessment](https://mlsanity.com/paper/2403.04963/) (2024)
- [Learning Human-Human Interactions in Images from Weak Textual Supervision](https://mlsanity.com/paper/2304.14104/) (2023)

## Abstract (arXiv)

In this work, we explore a useful but often neglected methodology for robustness analysis of text generation evaluation metrics: stress tests with synthetic data. Basically, we design and synthesize a wide range of potential errors and check whether they result in a commensurate drop in the metric scores. We examine a range of recently proposed evaluation metrics based on pretrained language models, for the tasks of open-ended generation, translation, and summarization. Our experiments reveal interesting insensitivities, biases, or even loopholes in existing metrics. For example, we find that BERTScore is confused by truncation errors in summarization, and MAUVE (built on top of GPT-2) is insensitive to errors at the beginning or middle of generations. Further, we investigate the reasons behind these blind spots and suggest practical workarounds for a more reliable evaluation of text generation. We have released our code and data at https://github.com/cloudygoose/blindspot_nlg.