Research map / Probabilistic and scientific modeling / Variational and generative sampling
Neural Variational Inference for Text Processing
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 |
|---|---|---|---|
| Replace deterministic attention with stochastic latent attention (NASM vs LSTM+Att) (ablation) | answer selection LSTM model · QASent | MAP: 0.6451 -> 0.6501 | Helped |
| Replace deterministic attention with stochastic latent attention (NASM vs LSTM+Att) (ablation) | answer selection LSTM model · WikiQA | MAP: 0.6639 -> 0.6705 | Helped |
| Add deterministic attention to vanilla LSTM (ablation) | answer selection LSTM model · WikiQA | MAP: 0.6552 -> 0.6639 | Helped |
| Add deterministic attention to vanilla LSTM (ablation) | answer selection LSTM model · QASent | MAP: 0.6436 -> 0.6451 | No clear effect |
| Add lexical overlap count feature to NASM (ablation) | NASM · QASent | MAP: 0.6501 -> 0.7339 | Helped |
| Add lexical overlap count feature to NASM (ablation) | NASM · WikiQA | MAP: 0.6705 -> 0.6886 | Helped |
| Use NVDM with continuous latent document representation instead of discrete topic models | document modelling · 20NewsGroups | perplexity: 836 vs best baseline 896 (docNADE) | Helped |
| Use NVDM with continuous latent document representation instead of discrete topic models | document modelling · RCV1-v2 | perplexity: 550 vs best baseline 598 (fDARN 200-dim) | Helped |
| Increase NVDM latent dimension from 50 to 200 (ablation) | NVDM · 20NewsGroups | perplexity: 836 -> 852 | Hurt |
| Increase NVDM latent dimension from 50 to 200 (ablation) | NVDM · RCV1-v2 | perplexity: 563 -> 550 | Helped |
Closest papers by meaning
- A Discrete Variational Recurrent Topic Model without the Reparametrization Trick (NeurIPS 2020 · similarity 0.90)
- Generating Well-Formed Answers by Machine Reading with Stochastic Selector Networks (AAAI 2020 · similarity 0.88)
- Variational Open-Domain Question Answering (ICML 2023 · similarity 0.88)
- Discovering Discrete Latent Topics with Neural Variational Inference (ICML 2017 · similarity 0.88)
- Latent Thought Models with Variational Bayes Inference-Time Computation (ICML 2025 · similarity 0.86)
- UniGen: A Unified Generative Framework for Retrieval and Question Answering with Large Language Models (AAAI 2024 · similarity 0.85)
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.