Research map / Learning theory and algorithms / Clustering
X-DMM: Fast and Scalable Model Based Text 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 |
|---|---|---|---|
| use Metropolis-Hastings sampling instead of collapsed Gibbs (ablation) | X-DMM · ohsumed K=240 | time per iteration (s): 23.5 -> 0.54 | Helped |
| use Metropolis-Hastings sampling instead of collapsed Gibbs (ablation) | X-DMM · ohsumed K=240 | iterations to converge: 48 -> 738 | Hurt |
| use symmetric instead of asymmetric Dirichlet prior (ablation) | X-DMM · 20ng, ohsumed, QA, Reuters | NMI | No clear effect |
| replace GSDMM with X-DMM | DMM text clustering · NYTimes perplexity 3881 | training time (s): 4162.5 -> 849.9 | Helped |
| add GPU parallel uncollapsed Gibbs sampler (ablation) | X-DMM · NYTimes perplexity 3881 | training time (s): 849.9 -> 262.3 | Helped |
| add GPU parallel uncollapsed Gibbs sampler (ablation) | X-DMM · 20ng K=40 | NMI: 0.691 -> 0.682 | Mixed |
| replace GSDMM with X-DMM | DMM text clustering · 4 text datasets | NMI: 0.689 -> 0.691 (20ng K=40) | No clear effect |
Closest papers by meaning
- Maximum Margin Dirichlet Process Mixtures for Clustering (AAAI 2016 · similarity 0.87)
- Optimized Algorithms for Text Clustering with LLM-Generated Constraints (AAAI 2026 · similarity 0.86)
- Summaries as Centroids for Interpretable and Scalable Text Clustering (ICLR 2026 · similarity 0.85)
- A Sampling-Based Approach for Efficient Clustering in Large Datasets (CVPR 2022 · similarity 0.84)
- Information-Theoretic Generative Clustering of Documents (AAAI 2025 · similarity 0.84)
- Nemotron-CLIMB: Clustering-based Iterative Data Mixture Bootstrapping for Language Model Pre-training (NeurIPS 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.