Research map / Probabilistic and scientific modeling / Dynamical systems
Latent LSTM Allocation: Joint Clustering and Non-Linear Dynamic Modeling of Sequence Data
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 |
|---|---|---|---|
| Joint training of topics and LSTM dynamics (vs. training LSTM on frozen LDA topic assignments) (ablation) | LLA · Wikipedia and User click history | perplexity: 2119 -> 1785 (Wikipedia); 1572 -> 927 (user clicks) | Helped |
| Modeling sequence order over topics (LLA) vs. LDA with 4x more topics (ablation) | LLA variants · Wikipedia | perplexity | Helped |
| Topic-level LSTM (LLA) instead of LDA bag-of-words model | LDA · User search click history | perplexity | Helped |
| Replace word embedding with char-level LSTM input in word LLA (char LLA) | LLA · Wikipedia | model size vs perplexity | Mixed |
| Char-level LSTM input instead of word lookup table in LSTM baseline | word-LSTM · Wikipedia | model size and perplexity | Mixed |
| Topic-level LSTM (LLA) vs ddLDA smoothness-based temporal LDA | LLA · User search click history | perplexity | Helped |
Closest papers by meaning
- Mogrifier LSTM (ICLR 2020 · similarity 0.87)
- User-Dependent Neural Sequence Models for Continuous-Time Event Data (NeurIPS 2020 · similarity 0.87)
- xLSTM: Extended Long Short-Term Memory (NeurIPS 2024 · similarity 0.85)
- Human-inspired Episodic Memory for Infinite Context LLMs (ICLR 2025 · similarity 0.85)
- Understanding and Improving Length Generalization in Recurrent Models (ICML 2025 · similarity 0.85)
- LLM Processes: Numerical Predictive Distributions Conditioned on Natural Language (NeurIPS 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.