Extending Item Response Theory for Efficient and Meaningful Multilingual Evaluation

• URL: http://arxiv.org/abs/2606.15643v1
• Date: Sun, 14 Jun 2026 07:16:03 GMT
• Title: Extending Item Response Theory for Efficient and Meaningful Multilingual Evaluation
• Abstract summary: Multilingual benchmarks are central to evaluating large language models (LLMs) across languages.<n>They suffer from three issues: exhaustive evaluation scales linearly with the number of languages, automatic translation introduces errors that are easily missed at scale, and some items conflate general and culture-specific knowledge.<n>We address all three with a unified statistical framework, Multilingual-IRT, which extends Item Response Theory with per-language difficulty deviations.
• Score: 19.965981631741894
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Multilingual benchmarks are central to evaluating large language models (LLMs) across languages, but they suffer from three issues: exhaustive evaluation scales linearly with the number of languages, automatic translation introduces errors that are easily missed at scale, and some items conflate general and culture-specific knowledge. We address all three with a unified statistical framework, Multilingual-IRT, which extends Item Response Theory with per-language difficulty deviations, split discriminability separating content from language effects, and per-language ability residuals. Fitting Multilingual-IRT on 25 LLMs across 29 languages of MMLU-Pro-X, we show that its fitted parameters support three practical applications: predicting unobserved (item, LLM, language) instances with 11-16% lower binary cross-entropy than the strongest accuracy-based baseline, surfacing candidate translation errors distributed across all 28 non-English languages, whereas accuracy-based baselines concentrate detections in a few languages, and recovering culture-specific items that accuracy-based baselines miss.

Related papers

• Lower-Resource, Higher Scores: Language Bias in LLM Evaluators [27.959960538188753]
  We conduct experiments with semantically identical instruction-response pairs across 23 languages.<n>We find that multilingual evaluators assign significantly different scores to different evaluation languages.
  arXiv  Detail & Related papers  (2026-07-16T01:53:09Z)
• The GaoYao Benchmark: A Comprehensive Framework for Evaluating Multilingual and Multicultural Abilities of Large Language Models [51.61416200800499]
  GaoYao is a comprehensive benchmark with 182.3k samples, 26 languages and 51 nations/areas.<n>First, GaoYao proposes a unified framework categorizing evaluation tasks into three cultural layers.<n>Second, we achieve native-quality expansion by leveraging experts to rigorously localize subjective benchmarks into 19 languages.<n>Third, we conduct an in-depth diagnostic analysis on 20+ flagship and compact LLMs.
  arXiv  Detail & Related papers  (2026-04-22T06:19:46Z)
• Calibrating Beyond English: Language Diversity for Better Quantized Multilingual LLM [10.689556615369272]
  Non-English and multilingual calibration sets significantly improve perplexity compared to English-only baselines.<n> tailoring calibration sets to the evaluation language yields the largest improvements for individual languages.
  arXiv  Detail & Related papers  (2026-01-26T09:36:03Z)
• M3TQA: Massively Multilingual Multitask Table Question Answering [39.99483693397598]
  m3TQA-Instruct is a large-scale benchmark spanning 97 languages across diverse language families.<n>We construct m3TQA by curating 50 real-world tables in Chinese and English, then applying a robust six-step translation pipeline powered by DeepSeek and GPT-4o.<n>The benchmark includes 2,916 professionally annotated question-answering pairs across four tasks designed to evaluate nuanced table reasoning capabilities.
  arXiv  Detail & Related papers  (2025-08-22T09:57:40Z)
• Learning the Topic, Not the Language: How LLMs Classify Online Immigration Discourse Across Languages [0.0]
  Large language models (LLMs) are transforming social-science research by enabling scalable, precise analysis.<n>We fine-tune lightweight LLaMA 3.2-3B models on monolingual, bilingual, or multilingual data sets to classify immigration-related tweets.<n>We evaluate whether minimal language-specific fine-tuning enables cross-lingual topic detection and whether adding targeted languages corrects pre-training biases.
  arXiv  Detail & Related papers  (2025-08-08T16:23:24Z)
• PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts [85.78821098963607]
  PolyMath is a multilingual mathematical reasoning benchmark covering 18 languages and 4 easy-to-hard difficulty levels.<n>Our benchmark ensures difficulty comprehensiveness, language diversity, and high-quality translation.
  arXiv  Detail & Related papers  (2025-04-25T15:39:04Z)
• MELA: Multilingual Evaluation of Linguistic Acceptability [7.524375463656369]
  We present the largest benchmark to date on linguistic acceptability: Multilingual Evaluation of Linguistic Acceptability -- MELA, with 46K samples covering 10 languages. In pursuit of multilingual interpretability, we conduct probing experiments with fine-tuned XLM-R. Cross-lingual transfer experiments show that transfer in acceptability judgment is non-trivial.
  arXiv  Detail & Related papers  (2023-11-15T15:25:28Z)
• Breaking Language Barriers in Multilingual Mathematical Reasoning: Insights and Observations [59.056367787688146]
  This paper pioneers exploring and training powerful Multilingual Math Reasoning (xMR) LLMs. We construct the first multilingual math reasoning instruction dataset, MGSM8KInstruct, encompassing ten distinct languages. By utilizing translation, we construct the first multilingual math reasoning instruction dataset, MGSM8KInstruct, encompassing ten distinct languages.
  arXiv  Detail & Related papers  (2023-10-31T08:09:20Z)
• Polyglot or Not? Measuring Multilingual Encyclopedic Knowledge in Foundation Models [0.0]
  We assess the ability of foundation models to recall encyclopedic knowledge across a wide range of linguistic contexts. We produce a 20-language dataset that contains 303k factual associations paired with counterfactuals. We evaluate 5 models in a multilingual test, and benchmark a diverse set of 24 models in an English-only test.
  arXiv  Detail & Related papers  (2023-05-23T04:31:39Z)
• AM2iCo: Evaluating Word Meaning in Context across Low-ResourceLanguages with Adversarial Examples [51.048234591165155]
  We present AM2iCo, Adversarial and Multilingual Meaning in Context. It aims to faithfully assess the ability of state-of-the-art (SotA) representation models to understand the identity of word meaning in cross-lingual contexts. Results reveal that current SotA pretrained encoders substantially lag behind human performance.
  arXiv  Detail & Related papers  (2021-04-17T20:23:45Z)
• Inducing Language-Agnostic Multilingual Representations [61.97381112847459]
  Cross-lingual representations have the potential to make NLP techniques available to the vast majority of languages in the world. We examine three approaches for this: (i) re-aligning the vector spaces of target languages to a pivot source language; (ii) removing language-specific means and variances, which yields better discriminativeness of embeddings as a by-product; and (iii) increasing input similarity across languages by removing morphological contractions and sentence reordering.
  arXiv  Detail & Related papers  (2020-08-20T17:58:56Z)
• Learning to Learn Morphological Inflection for Resource-Poor Languages [105.11499402984482]
  We propose to cast the task of morphological inflection - mapping a lemma to an indicated inflected form - for resource-poor languages as a meta-learning problem. Treating each language as a separate task, we use data from high-resource source languages to learn a set of model parameters. Experiments with two model architectures on 29 target languages from 3 families show that our suggested approach outperforms all baselines.
  arXiv  Detail & Related papers  (2020-04-28T05:13:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.