State Space Models
PulseAugur coverage of State Space Models — every cluster mentioning State Space Models across labs, papers, and developer communities, ranked by signal.
10 day(s) with sentiment data
-
New LFNet method fuses CNN and SSM features for improved salient object detection
Researchers have developed a novel method called Liquid Fusion Network (LFNet) to improve salient object detection by harmonizing features from different neural network architectures. LFNet addresses the spectral biases…
-
RotRNN: New Linear Recurrent Model Simplifies Long Sequence Modeling
Researchers have introduced RotRNN, a novel linear recurrent neural network designed for modeling long sequences. This model leverages rotation matrices to simplify initialization and normalization procedures, addressin…
-
New Hybrid Mamba-Transformer Model Enhances EHR Representation
Researchers have developed HyMaTE, a novel hybrid model that combines Mamba (a State Space Model) and Transformer architectures to improve the representation of electronic health records (EHRs). This approach aims to ov…
-
Lifelong AI Learning Needs Parametric Attention in Transformers, Paper Argues
A new research paper proposes that achieving lifelong continual learning in AI agents necessitates the use of parametric forms of attention within transformer models. The paper argues that the current quadratic complexi…
-
MambaRaw uses State Space Models for efficient 4K raw image reconstruction
Researchers have developed MambaRaw, a new framework for reconstructing high-resolution raw images using JPEG previews. This method leverages State Space Models (SSMs) to efficiently estimate entropy parameters, overcom…
-
Mamba models offer faster OCR but lag Transformer accuracy on historical texts
Researchers have benchmarked State-Space Models (SSMs), specifically Mamba, against Transformers and BiLSTMs for Optical Character Recognition (OCR) on historical newspapers. The studies indicate that while Mamba-based …
-
AI models adapt to new sensors and long-range data in remote sensing research
Two new arXiv papers explore advancements in applying machine learning to remote sensing data. The first paper surveys the use of State Space Models (SSMs) for tasks like dense visual predictions and temporal data analy…
-
Robotics research explores new self-supervision and SSMs for imitation learning · 2 sources tracked
Two new research papers explore advanced techniques for improving reinforcement learning in robotics. The first, Temporal Self-Imitation Learning (TSIL), introduces a method to use the temporal efficiency of successful …
-
New research explores diffusion and state space models beyond autoregressive AI
Two new arXiv papers explore advanced modeling techniques beyond traditional autoregressive language models. The first paper surveys Diffusion Models, Code World Models, and State Space Models for code intelligence, sug…
-
New research examines adversarial attacks on State-Space Models for robust RL
A new research paper explores how adversarial attacks can impact probabilistic State-Space Models (SSMs) used in reinforcement learning. The study analyzes how attackers can alter observations within likelihood constrai…
-
New research explores robust optimization and reinforcement learning techniques · 6 sources tracked
Several new research papers explore advanced techniques in reinforcement learning and optimization, focusing on robustness and generative models. One paper introduces a stationary robust mean-field game framework to add…
-
New framework reveals how State Space Models learn code, guides architectural improvements
Researchers have developed SSM-Interpret, a new framework for analyzing State Space Models (SSMs) used in code understanding. The study found that SSMs initially capture syntactic and semantic structures better than Tra…
-
New PRISMamba method enhances Vision SSMs with rotation robustness
Researchers have introduced PRISMamba, a novel approach to processing images within Vision State Space Models (SSMs). Unlike traditional methods that serialize images into linear sequences, PRISMamba partitions images i…
-
New library Dynestyx simplifies state-space models for machine learning
Researchers have introduced Dynestyx, a new probabilistic programming library designed to simplify the integration of state-space models (SSMs) into modern probabilistic programming languages. This library aims to make …
-
AI models advance respiratory sound classification with new techniques
Two new research papers propose advanced AI techniques for classifying respiratory sounds. One paper introduces QLung, a quality-adaptive framework that adjusts learning margins based on audio recording quality, improvi…
-
Blurry Window Attention improves Transformer efficiency for long contexts
Researchers have introduced Blurry Window Attention (BLA), a novel method designed to improve the efficiency of Transformer language models in handling long contexts. BLA addresses the quadratic complexity and growing K…
-
HiPPO Zoo enhances SSMs with explicit, interpretable memory
Researchers have introduced "HiPPO Zoo," a framework that enhances state space models (SSMs) by making their memory mechanisms explicit and interpretable. This approach builds upon the original HiPPO framework, which us…
-
GeoMag uses State Space Models for consistent video motion magnification
Researchers have developed GeoMag, a new framework for video motion magnification that utilizes State Space Models to enhance imperceptible dynamics while maintaining global structural consistency. This approach address…
-
New G-SLiCEs Model Advances Universal Time Series Generation
Researchers have introduced Generative SLiCEs (G-SLiCEs), a novel continuous-time model for generative time-series modeling. This model is built upon theoretical findings that maximally expressive Structured Linear Cont…
-
MedMamba architecture improves medical time series classification
Researchers have developed MedMamba, a novel architecture for medical time series classification that integrates state space models with adaptive graph learning. This approach aims to better capture local-global dynamic…