long short-term memory
PulseAugur coverage of long short-term memory — every cluster mentioning long short-term memory across labs, papers, and developer communities, ranked by signal.
- instance of DagsHub 90%
- instance of Gotit.pub 90%
- instance of ScienceCast 90%
- instance of alphaXiv 90%
- instance of Recurrent Neural Networks 90%
- developed by Recurrent Neural Networks 90%
- instance of BiLSTM 90%
- instance of N Beats Neural Basis Expansion Analysis For Time Series Forecasting 90%
- instance of Bi-LSTM-AMSM: bidirectional long short-term memory network and attention mechanism with semantic mining for e-commerce web page recommendation 90%
- competes with Mass-Conserving Perceptron 75%
- competes with gated recurrent unit 70%
- used by CatalyzeX 70%
- 2026-05-14 research_milestone A hybrid LSTM model achieved the lowest final displacement error in dynamic movement forecasting. source
17 day(s) with sentiment data
-
AI training optimizer cuts AllReduce latency by 36% on NVIDIA GPUs
Researchers have developed a novel multi-agent system to optimize collective communication libraries (CCL) for large-scale distributed AI/ML training. This system uses a Long Short-Term Memory (LSTM) network to detect n…
-
New two-stage residual learning method improves solar irradiance forecast uncertainty
Researchers have developed a novel two-stage residual learning method to improve the accuracy of solar irradiance forecasts, particularly in estimating uncertainty. This approach uses a first long short-term memory (LST…
-
Hybrid ML predicts rare aircraft landing safety events
Researchers have developed a hybrid machine learning framework to predict rare safety events during aircraft landing phases using imbalanced flight data. The system combines a long short-term memory (LSTM) autoencoder t…
-
Brain's spatial memory could enable universal computation, new paper suggests
A new paper proposes that finite brains can generalize without bound by utilizing spatial memory as a universal computer. The research suggests that current models like LSTMs and Transformers struggle with longer sequen…
-
New framework uses LSTM for granular construction material price forecasting
A new research paper introduces a granular framework for forecasting construction material prices, utilizing the Construction Specifications Institute (CSI) MasterFormat structure. The framework integrates explanatory v…
-
AI models improve sensor data forecasting and fault detection · 2 sources tracked
Researchers have developed methods to improve the accuracy and robustness of time-series forecasting models, particularly for sensor data. One study demonstrates that Long-Short Term Memory (LSTM) network ensembles can …
-
Process-based hydrological model matches deep learning accuracy
Researchers have developed a process-based hydrological model, named CaRT+, that achieves accuracy comparable to deep learning models on three continents. This new model, which incorporates physical laws like conservati…
-
New theory on RNN pruning via poset modeling published on arXiv
A new paper proposes a corrected theory for lattice-based pruning of recurrent neural networks (RNNs), focusing on how information flows between neurons. The research models RNN dependencies as a partially ordered set (…
-
Deep Learning Models Enhance Financial Fraud Detection
This paper introduces a deep learning approach to enhance financial cybersecurity by detecting fraudulent transactions. It utilizes neural network architectures, specifically long short-term memory (LSTM) and convolutio…
-
Deep learning architectures rigorously benchmarked for financial forecasting · 2 sources tracked
Two research papers explore the application and evaluation of deep learning models for financial forecasting. The first paper introduces FinForesight, a diagnostic benchmark designed to rigorously test deep research sys…
-
New LSTM-beta-VAE framework enhances AI for industrial anaerobic digestion
Researchers have developed a novel generative data augmentation framework using an LSTM-beta-Variational Autoencoder combined with a multi-output LSTM (MO-LSTM) soft sensor. This approach addresses data sparsity and the…
-
GateTS model simplifies time-series forecasting with efficient MoE architecture
Researchers have introduced GateTS, a novel architecture for time-series forecasting that simplifies training and enhances efficiency. This model utilizes a sparse Mixture-of-Experts (MoE) approach with an attention-ins…
-
LSTM-Style Transformers Reduce Parameters and Cache Needs
Researchers have developed an LSTM-style transformer that reuses weights and shares a key-value cache across its recurrent core, significantly reducing parameter count and cache requirements. This approach allows for de…
-
Deep learning models struggle with financial forecasting benchmarks, study finds
A new study published on arXiv evaluates 16 deep learning architectures for financial time series forecasting, finding that many popular models like generic Transformers and State Space Models underperform simpler bench…
-
WADE engine uses LSTM to detect WebAssembly application attacks
Researchers have developed WADE, a novel anomaly detection engine designed to identify malicious behaviors within WebAssembly applications. WADE utilizes a Long Short-Term Memory (LSTM) network to learn the normal execu…
-
Research questions memory advantage in AI agents, finds time step is key
A new research paper investigates the apparent memory advantage of LSTM agents over memory-free MLP agents in the XLand-MiniGrid task. The study found that a significant portion of the observed performance gap, up to 10…
-
DistilGPT-2 recommended for improving TinyGPT project
The author discusses various Large Language Models (LLMs) suitable for improving a project that previously used a small, custom-trained Transformer model called TinyGPT. They evaluate options like LSTM/GRU, DistilGPT-2/…
-
New methods enable quantization of Orthogonal Recurrent Neural Networks
Researchers have developed methods for quantizing Orthogonal Recurrent Neural Networks (ORNNs), creating Quantized approximately Orthogonal RNNs (QORNNs). This advancement addresses the challenge of deploying ORNNs on r…
-
PlayerEmbed framework learns behavioral fingerprints from raw game input
Researchers have developed PlayerEmbed, a novel framework designed to create compact and discriminative player embeddings from raw input sequences in video games. This system utilizes a hybrid CNN-LSTM encoder trained o…
-
StarRTT framework enhances RTT prediction for Starlink satellites
Researchers have developed StarRTT, a novel framework designed to improve Round-Trip Time (RTT) prediction in Starlink LEO satellite networks. This framework addresses the challenge of RTT fluctuations caused by frequen…