Data to 7 Oct 2026

Hierarchical Multiscale Recurrent Neural Networks

arXiv 1609.01704 ↗First posted 6 Sep 2016cs.LGcited by 102

Not on the map. Why?

Abstract (arXiv)

Learning both hierarchical and temporal representation has been among the long-standing challenges of recurrent neural networks. Multiscale recurrent neural networks have been considered as a promising approach to resolve this issue, yet there has been a lack of empirical evidence showing that this type of models can actually capture the temporal dependencies by discovering the latent hierarchical structure of the sequence. In this paper, we propose a novel multiscale approach, called the hierarchical multiscale recurrent neural networks, which can capture the latent hierarchical structure in the sequence by encoding the temporal dependencies with different timescales using a novel update mechanism. We show some evidence that our proposed multiscale architecture can discover underlying hierarchical structure in the sequences without using explicit boundary information. We evaluate our proposed model on character-level language modelling and handwriting sequence modelling.

Key facts. Cited by 102 papers in the map (data through 7 Oct 2026). First posted on arXiv 6 Sep 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 25; 2022: 28; 2023: 12; 2024: 16; 2025: 8; 2026: 13.

102 papers.

Sort:

By area:

2021ByT5: Towards a token-free future with pre-trained byte-to-byte modelscited by 504
2021Dynamic Neural Networks: A Surveycited by 405
2021CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representationcited by 229
2021PonderNet: Learning to Pondercited by 156
2021Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLPcited by 130
2023Hierarchically Gated Recurrent Neural Network for Sequence Modelingcited by 100
2022Efficient Transformers with Dynamic Token Poolingcited by 75
2024Simple Hierarchical Planning with Diffusioncited by 71
2022Scaleformer: Iterative Multi-scale Refining Transformers for Time Series Forecastingcited by 54
2022Long Document Summarization with Top-down and Bottom-up Inferencecited by 42
2021Clockwork Variational Autoencoderscited by 40
2022Transformer Grammars: Augmenting Transformer Language Models with Syntactic Inductive Biases at Scalecited by 40

38 more

2024Learning Iterative Reasoning through Energy Diffusioncited by 36
2022Learning Iterative Reasoning through Energy Minimizationcited by 32
2023Bridging Discrete and Backpropagation: Straight-Through and Beyondcited by 31
2022A Survey on Dynamic Neural Networks for Natural Language Processingcited by 24
2024Two Stones Hit One Bird: Bilevel Positional Encoding for Better Length Extrapolationcited by 23
2024Training LLMs over Neurally Compressed Textcited by 16
2021Learning Task Decomposition with Ordered Memory Policy Networkcited by 15
2022An Independently Learnable Hierarchical Model for Bilateral Control-Based Imitation Learning Applicationscited by 14
2022Time-multiplexed Neural Holography: A flexible framework for holographic near-eye displays with fast heavily-quantized spatial light modulatorscited by 14
2022Temporal Latent Bottleneck: Synthesis of Fast and Slow Processing Mechanisms in Sequence Learningcited by 13
2023BigSmall: Efficient Multi-Task Learning for Disparate Spatial and Temporal Physiological Measurementscited by 13
2021Hierarchical Recurrent Neural Networks for Conditional Melody Generation with Long-term Structurecited by 12
2022MSPred: Video Prediction at Multiple Spatio-Temporal Scales with Hierarchical Recurrent Networkscited by 12
2021Cluster-Promoting Quantization with Bit-Drop for Minimizing Network Quantization Losscited by 11
2024Natural Language Processing Methods for Symbolic Music Generation and Information Retrieval: a Surveycited by 10
2021A Masked Segmental Language Model for Unsupervised Natural Language Segmentationcited by 9
2021Variational Predictive Routing with Nested Subjective Timescalescited by 9
2021When in Doubt, Summon the Titans: Efficient Inference with Large Modelscited by 8
2022Training Discrete Deep Generative Models via Gapped Straight-Through Estimatorcited by 8
2022Multi-Horizon Representations with Hierarchical Forward Models for Reinforcement Learningcited by 7
2023A Block-Coordinate Approach of Multi-level Optimization with an Application to Physics-Informed Neural Networkscited by 7
2023SutraNets: Sub-series Autoregressive Networks for Long-Sequence, Probabilistic Forecastingcited by 7
2025ms-Mamba: Multi-scale Mamba for Time-Series Forecastingcited by 7
2024PlanDQ: Hierarchical Plan Orchestration via D-Conductor and Q-Performercited by 6
2021Unsupervised Automatic Speech Recognition: A Reviewcited by 5
2022Using Probabilistic Machine Learning to Better Model Temporal Patterns in Parameterizations: a case study with the Lorenz 96 modelcited by 5
2022MS-RNN: A Flexible Multi-Scale Framework for Spatiotemporal Predictive Learningcited by 5
2026Pretraining Recurrent Networks without Recurrencecited by 5
2021Integrating Approaches to Word Representationcited by 4
2021Recurrent Neural Networks for Learning Long-term Temporal Dependencies with Reanalysis of Time Scale Representationcited by 4
2022Hyperspherical Quantization: Toward Smaller and More Accurate Modelscited by 4
2022Long-horizon video prediction using a dynamic latent hierarchycited by 4
2023Selective Inference for Changepoint detection by Recurrent Neural Networkcited by 4
2024The Role of Temporal Hierarchy in Spiking Neural Networkscited by 4
2021Warming up recurrent neural networks to maximise reachable multistability greatly improves learningcited by 3
2021Two-Timescale End-to-End Learning for Channel Acquisition and Hybrid Precodingcited by 3
2023Hierarchical Attention Encoder Decodercited by 3
2024Deep Neural Networks are Adaptive to Function Regularity and Data Distribution in Approximation and Estimationcited by 3

50 most cited of 102 All 102 (JSON)