Research map / Label-efficient and robust learning / Recognition and long-tail
CSI: A Coarse Sense Inventory for 85% 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 |
|---|---|---|---|
| use CSI labels instead of WordNet senses for annotation | annotators · annotation task | Kraemer's kappa: 0.51 -> 0.81 | Helped |
| use CSI instead of SuperSenses for annotation | annotators · annotation task | Kraemer's kappa: 0.69 -> 0.81 | Helped |
| use CSI instead of WordNet Domains labels | annotators · descriptiveness | avg rank: 1.80 -> 2.23 | Helped |
| use CSI instead of SuperSenses | BERT + Dense WSD · ALL | F1: 81.8 -> 83.8 | Helped |
| use CSI instead of WordNet Domains | BERT + Dense WSD · ALL | F1: 87.2 -> 83.8 | Hurt |
| use CSI instead of WordNet Domains | BERT + Dense WSD · ALL | GTO: 1.36 -> 1.51 | Helped |
| use CSI instead of BabelDomains | BERT + Dense WSD · ALL | F1: 89.1 -> 91.1 | Helped |
| exclude semantically-empty label from valid answers (CSI vs WND) | WND · ALL | F1: 85.9 -> 89.9 | Helped |
| train on all SemCor instances covered by CSI | BERT + Dense CSI · ALL | F1: 83.8 -> 85.9 | Helped |
| use CSI instead of WND for few-shot (5 examples) gain over MFS | BERT + Dense · few-shot ALL | F1 gain over MFS: +5.7 vs +1.3 F1 | Helped |
| use CSI instead of SuperSenses for few-shot gain over MFS | BERT + Dense · few-shot ALL | F1 gain over MFS: +5.7 vs +4.3 F1 | Helped |
Closest papers by meaning
- Knowledge-based Word Sense Disambiguation using Topic Models (AAAI 2018 · similarity 0.90)
- XL-WSD: An Extra-Large and Cross-Lingual Evaluation Framework for Word Sense Disambiguation (AAAI 2021 · similarity 0.89)
- SensEmBERT: Context-Enhanced Sense Embeddings for Multilingual Word Sense Disambiguation (AAAI 2020 · similarity 0.88)
- Quantum Interference Model for Semantic Biases of Glosses in Word Sense Disambiguation (AAAI 2024 · similarity 0.86)
- Quantum-inspired Non-homologous Representation Constraint Mechanism for Long-tail Senses of Word Sense Disambiguation (AAAI 2025 · similarity 0.84)
- SoftMatcha: A Soft and Fast Pattern Matcher for Billion-Scale Corpus Searches (ICLR 2025 · similarity 0.83)
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.