Research map / Probabilistic and scientific modeling / Variational and generative sampling
Maximum Reconstruction Estimation for Generative Latent-Variable Models
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 MLE with maximum reconstruction estimation (MRE) | HMM, 50 states · PTB POS induction | many-to-one accuracy: 0.6108 -> 0.6739 | Helped |
| replace MLE with MRE | HMM, 50 states · PTB POS induction | variation of information: 3.2087 -> 2.7522 | Helped |
| replace MLE with MRE | HMM, 10 states · PTB POS induction | many-to-one accuracy: 0.4054 -> 0.3881 | Hurt |
| replace MLE with MRE | HMM, 20-40 states · PTB POS induction | many-to-one accuracy: 0.4804 -> 0.5203 | Helped |
| replace MLE with MRE | HMM, 10-50 states · PTB POS induction | variation of information: 3.0575 -> 2.9322 | Helped |
| replace MLE with MRE, small training set (10,000 sentences) | HMM · PTB POS induction | many-to-one accuracy: 0.5087 -> 0.5825 | Helped |
| replace MLE with MRE, 40,000 training sentences | HMM · PTB POS induction | many-to-one accuracy: 0.5800 -> 0.6112 | Helped |
| replace MLE with MRE, 10,000 sentences | HMM · PTB POS induction | variation of information: 3.3471 -> 2.9018 | Helped |
| replace MLE with MRE | IBM Model 2, C->E · FBIS word alignment | AER: 40.28 -> 38.33 | Helped |
| replace MLE with MRE | IBM Model 1, C->E · FBIS word alignment | AER: 43.07 -> 41.90 | Helped |
| replace MLE with MRE | IBM Model 1, E->C · FBIS word alignment | AER: 45.89 -> 45.39 | Helped |
| replace MLE with MRE | IBM Model 2, E->C · FBIS word alignment | AER: 42.38 -> 41.73 | Helped |
Closest papers by meaning
- MBR and QE Finetuning: Training-time Distillation of the Best and Most Expensive Decoding Methods (ICLR 2024 · similarity 0.85)
- Latent Thought Models with Variational Bayes Inference-Time Computation (ICML 2025 · similarity 0.85)
- A Bias-Variance-Covariance Decomposition of Kernel Scores for Generative Models (ICML 2024 · similarity 0.85)
- Continuous Autoregressive Modeling with Stochastic Monotonic Alignment for Speech Synthesis (ICLR 2025 · similarity 0.84)
- A Probabilistic Perspective on Unlearning and Alignment for Large Language Models (ICLR 2025 · similarity 0.84)
- Training-Free Bayesianization for Low-Rank Adapters of Large Language Models (NeurIPS 2025 · 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.