# JuICE: A Benchmark for Evaluating LLM-Judge in Identifying Cultural Errors

arXiv: https://arxiv.org/abs/2605.26955 · HTML: https://mlsanity.com/paper/2605.26955/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 35 references that are in the map.
- First posted on arXiv 26 May 2026.

## Builds on (most cited first, 23 of 23)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judge](https://mlsanity.com/paper/2411.16594/) (2024)
- [Humans or LLMs as the Judge? A Study on Judgement Biases](https://mlsanity.com/paper/2402.10669/) (2024)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages](https://mlsanity.com/paper/2406.09948/) (2024)
- [Having Beer after Prayer? Measuring Cultural Bias in Large Language Models](https://mlsanity.com/paper/2305.14456/) (2023)
- [VERISCORE: Evaluating the factuality of verifiable claims in long-form text generation](https://mlsanity.com/paper/2406.19276/) (2024)
- [Survey of Cultural Awareness in Language Models: Text and Beyond](https://mlsanity.com/paper/2411.00860/) (2024)
- [Massively Multi-Cultural Knowledge Acquisition & LM Benchmarking](https://mlsanity.com/paper/2402.09369/) (2024)
- [Culture is Not Trivia: Sociocultural Theory for Cultural NLP](https://mlsanity.com/paper/2502.12057/) (2025)
- [MM-Eval: A Multilingual Meta-Evaluation Benchmark for LLM-as-a-Judge and Reward Models](https://mlsanity.com/paper/2410.17578/) (2024)
- [CaLMQA: Exploring culturally specific long-form question answering across 23 languages](https://mlsanity.com/paper/2406.17761/) (2024)
- [LLM-GLOBE: A Benchmark Evaluating the Cultural Values Embedded in LLM Output](https://mlsanity.com/paper/2411.06032/) (2024)
- [Extrinsic Evaluation of Cultural Competence in Large Language Models](https://mlsanity.com/paper/2406.11565/) (2024)
- [The Case for "Thick Evaluations" of Cultural Representation in AI](https://mlsanity.com/paper/2503.19075/) (2025)
- [Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theory](https://mlsanity.com/paper/2602.00521/) (2026)
- [MAKIEval: A Multilingual Automatic WiKidata-based Framework for Cultural Awareness Evaluation for LLMs](https://mlsanity.com/paper/2505.21693/) (2025)
- [Culture is Everywhere: A Call for Intentionally Cultural Evaluation](https://mlsanity.com/paper/2509.01301/) (2025)
- [TALES: A Taxonomy and Analysis of Cultural Representations in LLM-generated Stories](https://mlsanity.com/paper/2511.21322/) (2025)
- [LLMs as Span Annotators: A Comparative Study of LLMs and Humans](https://mlsanity.com/paper/2504.08697/) (2025)
- [CURE: Cultural Understanding and Reasoning Evaluation - A Framework for "Thick" Culture Alignment Evaluation in LLMs](https://mlsanity.com/paper/2511.12014/) (2025)
- [Localizing and Mitigating Errors in Long-form Question Answering](https://mlsanity.com/paper/2407.11930/) (2024)

## Abstract (arXiv)

As large language models (LLMs) are increasingly deployed to users around the world, they are integrated into everyday tasks across diverse cultural contexts, from drafting personal communications to brainstorming creative ideas. These tasks are inherently cultural: they require contextual appropriateness, symbolic resonance, and tacit cultural expectations that native speakers draw on instinctively, meaning that a response can be factually plausible yet unmistakably wrong to a local reader. Existing cultural benchmarks have treated culture as a flat set of facts via fact verification or norm entailment methods, and have adopted LLM-as-a-Judge without examining whether they can capture such thick cultural errors. To address this gap, we present JuICE (Benchmark for LLM-Judge in Identifying Cultural Errors), a multilingual dataset of 7,470 span-level annotations of cultural and linguistic errors in long-form LLM responses. It covers 1,050 query-response pairs from four countries (the United States, South Korea, Indonesia, and Bangladesh), in both English and their countries' main languages. Using JuICE, we find that even the strongest LLM-judge achieves only an F1 of 0.52 in the erroneous span detection task. Furthermore, LLM-judges consistently miss thick cultural errors that local residents readily identify. Our findings suggest that robust cultural evaluation must move beyond surface-level detection toward frameworks that account for the depth and situatedness of cultural meaning.