{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/one-ruler-to-measure-them-all-benchmarking","title":"One ruler to measure them all: Benchmarking multilingual long-context language models","arxiv_id":"2503.01996","date":"2025-03-03","proceeding":null,"authors":["Yekyung Kim","Jenna Russell","Marzena Karpinska","Mohit Iyyer"],"abstract":"We present ONERULER, a multilingual benchmark designed to evaluate long-context language models across 26 languages. ONERULER adapts the English-only RULER benchmark (Hsieh et al., 2024) by including seven synthetic tasks that test both retrieval and aggregation, including new variations of the \"needle-in-a-haystack\" task that allow for the possibility of a nonexistent needle. We create ONERULER through a two-step process, first writing English instructions for each task and then collaborating with native speakers to translate them into 25 additional languages. Experiments with both open-weight and closed LLMs reveal a widening performance gap between low- and high-resource languages as context length increases from 8K to 128K tokens. Surprisingly, English is not the top-performing language on long-context tasks (ranked 6th out of 26), with Polish emerging as the top language. Our experiments also show that many LLMs (particularly OpenAI's o3-mini-high) incorrectly predict the absence of an answer, even in high-resource languages. Finally, in cross-lingual scenarios where instructions and context appear in different languages, performance can fluctuate by up to 20% depending on the instruction language. We hope the release of ONERULER will facilitate future research into improving multilingual and cross-lingual long-context training pipelines.","url_abs":"https://arxiv.org/abs/2503.01996v1","url_pdf":"https://arxiv.org/pdf/2503.01996v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"one-ruler-to-measure-them-all-benchmarking","repo_url":"https://github.com/mungg/OneRuler","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":null,"task_name":"8k"},{"task_slug":"all","task_name":"All"},{"task_slug":"benchmarking","task_name":"Benchmarking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.01996","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}