LLM Digest

Story

arxiv_llm_reliability · Oct 28, 2026 · paper

Source brief

Gold vs.~Translation: Are We Measuring Language, Knowledge, or Locality in Translated Benchmarks?

arxiv.orgOct 28, 2026
original source linked

In brief

Translated versions of English-origin benchmarks such as MMLU are widely used to evaluate large language models in non-English languages, yet it remains unclear whether they measure the same constructs as natively created benchmarks. We frame this as a proxy validity question and

Continues in

Large Language — open the evidence trace →

Read the original at arxiv.org →Open in live feed Read that day’s brief

Gold vs.~Translation: Are We Measuring Language, Knowledge, or Locality in Translated Benchmarks?

Earlier in this thread 4 items

Automated Materials Data and Knowledge Extraction using Machine Learning and Large Language Models

Hallucinations of Large Language Models in Cross-Language Code Translation: Phenomena and Mitigation

Seq@K: Evaluating Language Models Under Sequential Feedback

Reva: Real-Time Vulnerability Detection through Contextual Probing of Large Language Models