AWhat changed
Combined multiple classification, structured-prediction, question-answering and retrieval tasks across languages; the revision added harder multilingual-pool retrieval, broader language coverage and behavioral diagnostics.
Scientific scope
Combined multiple classification, structured-prediction, question-answering and retrieval tasks across languages; the revision added harder multilingual-pool retrieval, broader language coverage and behavioral diagnostics. Task and language coverage is uneven, not a full Cartesian product. Constituent datasets keep their original creators. XTREME-R modifies fine-tuning/evaluation for some retained tasks and is not directly score-equivalent to XTREME. Leaderboards compare models plus differing training data, not architecture alone.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Task and language coverage is uneven, not a full Cartesian product. Constituent datasets keep their original creators. XTREME-R modifies fine-tuning/evaluation for some retained tasks and is not directly score-equivalent to XTREME. Leaderboards compare models plus differing training data, not architecture alone.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization
- XTREME-R: Towards More Challenging and Nuanced Multilingual Evaluation
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction