# HalluMix: A Task-Agnostic, Multi-Domain Benchmark for Real-World Hallucination Detection

arXiv: https://arxiv.org/abs/2505.00506 · HTML: https://mlsanity.com/paper/2505.00506/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 26 references that are in the map.
- First posted on arXiv 1 May 2025.

## Builds on (most cited first, 18 of 18)

- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [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)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [Know What You Don't Know: Unanswerable Questions for SQuAD](https://mlsanity.com/paper/1806.03822/) (2018)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [Get To The Point: Summarization with Pointer-Generator Networks](https://mlsanity.com/paper/1704.04368/) (2017)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs](https://mlsanity.com/paper/1903.00161/) (2019)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [The NarrativeQA Reading Comprehension Challenge](https://mlsanity.com/paper/1712.07040/) (2017)
- [A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents](https://mlsanity.com/paper/1804.05685/) (2018)
- [Efficient Attentions for Long Document Summarization](https://mlsanity.com/paper/2104.02112/) (2021)
- [RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models](https://mlsanity.com/paper/2401.00396/) (2023)
- [MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documents](https://mlsanity.com/paper/2404.10774/) (2024)
- [DialogSum: A Real-Life Scenario Dialogue Summarization Dataset](https://mlsanity.com/paper/2105.06762/) (2021)
- [Lynx: An Open Source Hallucination Evaluation Model](https://mlsanity.com/paper/2407.08488/) (2024)

## Built on by (most cited first, 3 of 3)

- [PRISM: Probing Reasoning, Instruction, and Source Memory in LLM Hallucinations](https://mlsanity.com/paper/2604.16909/) (2026)
- [OpenHalDet: A Unified Benchmark for Hallucination Detection across Diverse Generation Scenarios](https://mlsanity.com/paper/2606.06959/) (2026)
- [AtomCite: Verification and Correction of Supplied Page-Level Citations in Multi-Page Documents](https://mlsanity.com/paper/2609.05802/) (2026)

## Abstract (arXiv)

As large language models (LLMs) are increasingly deployed in high-stakes domains, detecting hallucinated contentunicodex2013text that is not grounded in supporting evidenceunicodex2013has become a critical challenge. Existing benchmarks for hallucination detection are often synthetically generated, narrowly focused on extractive question answering, and fail to capture the complexity of real-world scenarios involving multi-document contexts and full-sentence outputs. We introduce the HalluMix Benchmark, a diverse, task-agnostic dataset that includes examples from a range of domains and formats. Using this benchmark, we evaluate seven hallucination detection systemsunicodex2013both open and closed sourceunicodex2013highlighting differences in performance across tasks, document lengths, and input representations. Our analysis highlights substantial performance disparities between short and long contexts, with critical implications for real-world Retrieval Augmented Generation (RAG) implementations. Quotient Detections achieves the best overall performance, with an accuracy of 0.82 and an F1 score of 0.84.