# "Knowing When You Don't Know": A Multilingual Relevance Assessment Dataset for Robust Retrieval-Augmented Generation

arXiv: https://arxiv.org/abs/2312.11361 · HTML: https://mlsanity.com/paper/2312.11361/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in RAG robustness and adaptive retrieval, one of 15 directions in Retrieval-augmented generation.
- Builds on 10 of its 58 references that are in the map.
- First posted on arXiv 18 Dec 2023.

## Placed in

- Retrieval-augmented generation / [RAG robustness and adaptive retrieval](https://mlsanity.com/rag/rag-robustness-and-adaptive-retrieval/)

## Builds on (most cited first, 10 of 10)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [How Close is ChatGPT to Human Experts? Comparison Corpus, Evaluation, and Detection](https://mlsanity.com/paper/2301.07597/) (2023)
- [Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Models](https://mlsanity.com/paper/2311.09210/) (2023)
- [Orca 2: Teaching Small Language Models How to Reason](https://mlsanity.com/paper/2311.11045/) (2023)
- [Rethinking with Retrieval: Faithful Large Language Model Inference](https://mlsanity.com/paper/2301.00303/) (2022)
- [Towards Best Practices for Training Multilingual Dense Retrieval Models](https://mlsanity.com/paper/2204.02363/) (2022)

## Built on by (most cited first, 13 of 13)

- [Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaboration](https://mlsanity.com/paper/2402.00367/) (2024)
- [Enhancing Noise Robustness of Retrieval-Augmented Language Models with Adaptive Adversarial Training](https://mlsanity.com/paper/2405.20978/) (2024)
- [A Comprehensive Survey of Retrieval-Augmented Generation (RAG): Evolution, Current Landscape and Future Directions](https://mlsanity.com/paper/2410.12837/) (2024)
- [Multilingual Large Language Model: A Survey of Resources, Taxonomy and Frontiers](https://mlsanity.com/paper/2404.04925/) (2024)
- [Retrieval-augmented generation in multilingual settings](https://mlsanity.com/paper/2407.01463/) (2024)
- [MIRAGE-Bench: Automatic Multilingual Benchmark Arena for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2410.13716/) (2024)
- [XRAG: Cross-lingual Retrieval-Augmented Generation](https://mlsanity.com/paper/2505.10089/) (2025)
- [Trustworthy Machine Learning via Memorization and the Granular Long-Tail: A Survey on Interactions, Tradeoffs, and Beyond](https://mlsanity.com/paper/2503.07501/) (2025)
- [Evaluating the Retrieval Robustness of Large Language Models](https://mlsanity.com/paper/2505.21870/) (2025)
- [Magic Mushroom: A Customizable Benchmark for Fine-grained Analysis of Retrieval Noise Erosion in RAG Systems](https://mlsanity.com/paper/2506.03901/) (2025)
- [Challenges and Recommendations for LLM-as-a-Judge in Multilingual Settings and for Low-Resource Languages](https://mlsanity.com/paper/2607.02235/) (2026)
- [Wontopos Tablet 2: Measuring Multilingual and Multimodal Memory Retrieval Without Lexical Matching](https://mlsanity.com/paper/2608.23920/) (2026)
- [RAGStress: A controlled benchmark for evaluating retrieval-augmented generation under knowledge-base degradation](https://mlsanity.com/paper/2610.04691/) (2026)

## Abstract (arXiv)

Retrieval-Augmented Generation (RAG) grounds Large Language Model (LLM) output by leveraging external knowledge sources to reduce factual hallucinations. However, prior work lacks a comprehensive evaluation of different language families, making it challenging to evaluate LLM robustness against errors in external retrieved knowledge. To overcome this, we establish NoMIRACL, a human-annotated dataset for evaluating LLM robustness in RAG across 18 typologically diverse languages. NoMIRACL includes both a non-relevant and a relevant subset. Queries in the non-relevant subset contain passages judged as non-relevant, whereas queries in the relevant subset include at least a single judged relevant passage. We measure relevance assessment using: (i) hallucination rate, measuring model tendency to hallucinate, when the answer is not present in passages in the non-relevant subset, and (ii) error rate, measuring model inaccuracy to recognize relevant passages in the relevant subset.In our work, we observe that most models struggle to balance the two capacities. Models such as LLAMA-2 and Orca-2 achieve over 88% hallucination rate on the non-relevant subset. Mistral and LLAMA-3 hallucinate less but can achieve up to a 74.9% error rate on the relevant subset. Overall, GPT-4 is observed to provide the best tradeoff on both subsets, highlighting future work necessary to improve LLM robustness. NoMIRACL dataset and evaluation code are available at: https://github.com/project-miracl/nomiracl.