# Reassessing Extractive QA Datasets at Scale: LLM-as-a-Judge and In-Depth Analyses

arXiv: https://arxiv.org/abs/2504.11972 · HTML: https://mlsanity.com/paper/2504.11972/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Builds on 36 of its 48 references that are in the map.
- First posted on arXiv 16 Apr 2025.

## Builds on (most cited first, 30 of 36)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Steps](https://mlsanity.com/paper/2011.01060/) (2020)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Teaching Machines to Read and Comprehend](https://mlsanity.com/paper/1506.03340/) (2015)
- [DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs](https://mlsanity.com/paper/1903.00161/) (2019)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies](https://mlsanity.com/paper/2101.02235/) (2021)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [CoQA: A Conversational Question Answering Challenge](https://mlsanity.com/paper/1808.07042/) (2018)
- [LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methods](https://mlsanity.com/paper/2412.05579/) (2024)
- [Large Language Models Are State-of-the-Art Evaluators of Translation Quality](https://mlsanity.com/paper/2302.14520/) (2023)
- [QuAC : Question Answering in Context](https://mlsanity.com/paper/1808.07036/) (2018)
- [NewsQA: A Machine Comprehension Dataset](https://mlsanity.com/paper/1611.09830/) (2016)
- [Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models](https://mlsanity.com/paper/2404.18796/) (2024)
- [Constructing Datasets for Multi-hop Reading Comprehension Across Documents](https://mlsanity.com/paper/1710.06481/) (2017)
- [Evaluating Correctness and Faithfulness of Instruction-Following Models for Question Answering](https://mlsanity.com/paper/2307.16877/) (2023)
- [Optimization-based Prompt Injection Attack to LLM-as-a-Judge](https://mlsanity.com/paper/2403.17710/) (2024)
- [Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation](https://mlsanity.com/paper/2212.07981/) (2022)
- [Does Prompt Formatting Have Any Impact on LLM Performance?](https://mlsanity.com/paper/2411.10541/) (2024)
- [Is LLM-as-a-Judge Robust? Investigating Universal Adversarial Attacks on Zero-shot LLM Assessment](https://mlsanity.com/paper/2402.14016/) (2024)

## Built on by (most cited first, 14 of 14)

- [Answer Matching Outperforms Multiple Choice for Language Model Evaluation](https://mlsanity.com/paper/2507.02856/) (2025)
- [Automatically Benchmarking LLM Code Agents through Agent-Driven Annotation and Evaluation](https://mlsanity.com/paper/2510.24358/) (2025)
- [Garbage In, Reasoning Out? Why Benchmark Scores are Unreliable and What to Do About It](https://mlsanity.com/paper/2506.23864/) (2025)
- [Watch, Listen, Understand, Mislead: Tri-modal Adversarial Attacks on Short Videos for Content Appropriateness Evaluation](https://mlsanity.com/paper/2507.11968/) (2025)
- [Pioneer Agent: Continual Improvement of Small Language Models in Production](https://mlsanity.com/paper/2604.09791/) (2026)
- [AirQA: A Comprehensive QA Dataset for AI Research with Instance-Level Evaluation](https://mlsanity.com/paper/2509.16952/) (2025)
- [Understanding Parametric Knowledge Injection in Retrieval-Augmented Generation](https://mlsanity.com/paper/2510.12668/) (2025)
- [Who Gets Left Behind? Auditing Disability Inclusivity in Large Language Models](https://mlsanity.com/paper/2509.00963/) (2025)
- [EnviroLLM: Resource Tracking and Optimization for Local AI](https://mlsanity.com/paper/2512.12004/) (2025)
- [Exploring the features used for summary evaluation by Human and GPT](https://mlsanity.com/paper/2512.19620/) (2025)
- [Error Taxonomy-Guided Prompt Optimization](https://mlsanity.com/paper/2602.00997/) (2026)
- [Structure-Aware RAG: Structured Retrieval Augmented Generation from Noisy Data for Conversational Agents](https://mlsanity.com/paper/2605.24366/) (2026)
- [What Do Rationales Communicate? A Message-Intervention Study in Role-Specialized QA](https://mlsanity.com/paper/2610.00018/) (2026)
- [Investigating the Role of Reasoning-Language Alignment in Monolingual Retrieval-Augmented Generation](https://mlsanity.com/paper/2610.03136/) (2026)

## Abstract (arXiv)

Extractive QA tasks are commonly evaluated using Exact Match (EM) and F1-score, but these metrics often fail to reflect true model performance. Recent studies have proposed using large language models (LLMs) as judges (LLM-as-a-judge), yet they often lack comprehensive evaluation across datasets and overlook key factors such as sensitivity to answer types, prompt variations, and self-preference bias. In this work, we conduct a systematic study of LLM-as-a-judge across four extractive QA datasets and various prompt variations, assessing multiple LLM families in both answering and judging roles. Our results show that LLM-as-a-judge judgments correlate much more strongly with human evaluations than EM (0.22) and F1 (0.40), achieving correlations up to 0.85 with open-source models. Further analysis reveals that LLM-as-a-judge performs particularly well on number-related answers but faces challenges with more complex types, such as job titles. Contrary to findings in other NLP tasks, we observe no self-preference bias, even when the same model serves as both QA model and judge. Finally, we find that prompt phrasing has minimal impact, and zero-shot, context-free judging often yields the best evaluation performance.