Research map / Learning theory and algorithms / Clustering
Bayesian Verb Sense Clustering
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 |
|---|---|---|---|
| PPMI clustering with pattern features instead of slot features (ablation) | PPMI mixture · Gigaword verb clustering | F1: 49.8 ± 1.6 -> 52.7 ± 0.2 | Helped |
| pattern features instead of slot features (ablation) | Dirichlet-Multinomial · Gigaword verb clustering | F1: 48.2 ± 1.5 -> 41.4 ± 1.8 | Hurt |
| pattern features instead of slot features (ablation) | PPMI mixture · Google Books verb clustering | F1: 24.07 -> 29.18 | Helped |
| pattern features instead of slot features (ablation) | Dirichlet-Multinomial · Google Books verb clustering | F1: 36.05 -> 36.49 | Helped |
| replace D-M with PPMI second-step clustering (pattern features) | verb clustering · Gigaword | F1: 41.4 ± 1.8 -> 52.7 ± 0.2 | Helped |
| replace D-M with PPMI second-step clustering (slot features) | verb clustering · Gigaword | F1: 48.2 ± 1.5 -> 49.8 ± 1.6 | Helped |
| replace D-M with PPMI second-step clustering (slot features) | verb clustering · Google Books | F1: 36.05 -> 24.07 | Hurt |
| replace D-M with PPMI second-step clustering (pattern features) | verb clustering · Google Books | F1: 36.49 -> 29.18 | Hurt |
| modified LDA sense induction versus prior published model | sense induction · Gigaword (micro) | F1: 32.96 -> 56.37 | Helped |
| sense induction with 100 topics instead of 200 | mLDA · Gigaword (micro) | F1: 53.95 -> 56.37 | Helped |
| Google Books instead of Gigaword corpus for sense induction | mLDA-100 · micro F1 | F1: 56.37 -> 40.97 | Hurt |
| PPMI instead of Dirichlet-Multinomial clustering, runtime | verb clustering · Gigaword-100 slots | runtime (s): 7400 -> 160 | Helped |
Closest papers by meaning
- Verb Class Induction with Partial Supervision (AAAI 2020 · similarity 0.92)
- Paloma: A Benchmark for Evaluating Language Model Fit (NeurIPS 2024 · similarity 0.84)
- Bayesian Test-Time Inference of Task-Aligned Similarity from Weak Interactive Feedback (NeurIPS 2026 · similarity 0.83)
- Language Model Cascades: Token-Level Uncertainty And Beyond (ICLR 2024 · similarity 0.83)
- Nemotron-CLIMB: Clustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training (NeurIPS 2025 · similarity 0.83)
- Data Mixture Inference Attack: BPE Tokenizers Reveal Training Data Compositions (NeurIPS 2024 · 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.