Metraipolis › archive › dc_a6ebbcddf096
XHotpotQA: A Benchmark for Cross-Lingual Knowledge Composition in Multi-Hop Question Answering — arXiv:2608.27481
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.27481
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T10:47:45.892Z
- SHA-256 of the archived text
f89e622f1dece120ddefb63243f55e752e29446cfef0477815c2fb11cdbda953- Stable URL of this page
https://metraipolis.com/documento/dc_a6ebbcddf096
How to cite
XHotpotQA: A Benchmark for Cross-Lingual Knowledge Composition in Multi-Hop Question Answering — arXiv:2608.27481 — arXiv (en), https://arxiv.org/abs/2608.27481 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a6ebbcddf096 (Metraipolis archive dc_a6ebbcddf096, sha256 f89e622f1dece120ddefb63243f55e752e29446cfef0477815c2fb11cdbda953)
BibTeX
@misc{dca6ebbcddf096,
title = {XHotpotQA: A Benchmark for Cross-Lingual Knowledge Composition in Multi-Hop Question Answering — arXiv:2608.27481},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.27481},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a6ebbcddf096; sha256 f89e622f1dece120ddefb63243f55e752e29446cfef0477815c2fb11cdbda953},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27481. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Knowledge-intensive multi-hop question answering requires systems to select evidence and compose dependent facts, yet multilingual benchmarks usually translate an entire example into one language. This hides failures at language boundaries inside the reasoning chain. We introduce XHotpotQA, a controlled benchmark for cross-lingual knowledge composition over mixed-language evidence. Each instance is modeled as an evidence-dependency graph whose question, bridge evidence, answer-bearing evidence, and distractors have explicit language assignments. The audited resource contains 15,661 training and 7,405 validation instances, with sentence-level support supervision and supplied distractors. In validation, 99.81% of items cross the question-to-gold-evidence language interface and 95.60% use gold paragraphs in different languages. Across three reader artifacts, full question-evidence mismatch is associated with 10.25 to 15.79 lower Unicode-aware answer F1 than partial alignment, and different-script evidence with deficits of 11.98 to 23.70 points; the corresponding adapted-selector contrasts are 1.71 and 1.78 points. Under this supplied-candidate design, the evaluated readers therefore show substantially larger condition-associated deficits than the selector. XHotpotQA provides role-aware diagnostics, modular evaluation, and an audited test bed for knowledge-based systems that must integrate evidence across languages. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.27481 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T10:33:42.444Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).