Research map / Large language models / Language, safety and interpretability
XL-WSD: An Extra-Large and Cross-Lingual Evaluation Framework for Word Sense Disambiguation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Train multilingual XLMR-L on English SC+WNG rather than language-specific data | XLMR-L WSD · Non-English XL-WSD test sets | F1: zero-shot XLMR-L: 47.15–83.88; language-specific XLMR-L: 36.00–78.67 | Mixed |
| Train on translated SemCor and WordNet Gloss rather than MULAN data | Language-specific XLMR-L · Italian | F1: 74.10 -> 77.70 | Helped |
| Train on translated SemCor and WordNet Gloss rather than MULAN data | Language-specific XLMR-L · Spanish | F1: 73.47 -> 77.26 | Helped |
| Use zero-shot XLMR-L rather than the strongest reported knowledge-based approach | WSD on non-English XL-WSD test sets · Non-English languages | F1: 65.66 vs 57.68 | Helped |
| Use XLMR-L rather than XLMR-B in the zero-shot setting | WSD across languages · XL-WSD test sets | F1: 65.66 vs 64.82 | Mixed |
| Use English training data and a multilingual model for cross-lingual transfer | XLMR-L WSD · Basque, Catalan, and Chinese | F1: 47.15; 49.97; 51.62 | Hurt |
Closest papers by meaning
- Language Ranker: A Metric for Quantifying LLM Performance Across High and Low-Resource Languages (AAAI 2025 · similarity 0.85)
- Quantum Interference Model for Semantic Biases of Glosses in Word Sense Disambiguation (AAAI 2024 · similarity 0.85)
- Quantum-inspired Non-homologous Representation Constraint Mechanism for Long-tail Senses of Word Sense Disambiguation (AAAI 2025 · similarity 0.85)
- Soft Token Alignment for Cross-Lingual Reasoning (NeurIPS 2026 · similarity 0.85)
- INCLUDE: Evaluating Multilingual Language Understanding with Regional Knowledge (ICLR 2025 · similarity 0.84)
- metabench - A Sparse Benchmark of Reasoning and Knowledge in Large Language Models (ICLR 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.