Research map / Probabilistic and scientific modeling / Variational and generative sampling
On-the-fly Rectification for Robust Large-Vocabulary Topic Inference
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 ENN with LAW instead of AP with AW | Topic inference · NeurIPS, New York Times, Movies, Songs | topic quality | No clear effect |
| Use LR-JSMF instead of AP+AW | Topic inference · NeurIPS, New York Times, Movies, Songs | runtime: 1∼2 orders of magnitude | Helped |
| Use PALM+LAW instead of AP+AW | Topic inference · NeurIPS, New York Times, Movies, Songs | topic quality | Mixed |
| Increase vocabulary size | LR-JSMF · NeurIPS | relative recovery error | Helped |
| Increase vocabulary size | LR-JSMF · NeurIPS | MST-Incoherence | Helped |
| Initialize ENN from the raw word-document data using randomized eigendecomposition | LR-JSMF · NeurIPS, New York Times, Movies, Songs | topic quality | No clear effect |
Closest papers by meaning
- A Discrete Variational Recurrent Topic Model without the Reparametrization Trick (NeurIPS 2020 · similarity 0.85)
- Precision-Recall Balanced Topic Modelling (NeurIPS 2019 · similarity 0.85)
- Knowledge-Aware Bayesian Deep Topic Model (NeurIPS 2022 · similarity 0.85)
- Provable Algorithms for Inference in Topic Models (ICML 2016 · similarity 0.85)
- Anchor & Transform: Learning Sparse Embeddings for Large Vocabularies (ICLR 2021 · similarity 0.84)
- Co-occurrence is not Factual Association in Language Models (NeurIPS 2024 · 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.