AINeutralarXiv – CS AI · May 126/10
🧠Researchers introduce FRACTAL, a novel state space model architecture that integrates fractional measure theory to improve long-sequence modeling by balancing short-term sensitivity with long-term memory retention. The approach achieves 87.11% on the Long Range Arena benchmark, outperforming existing SSM models like S5, addressing a fundamental trade-off in temporal sequence analysis.
AINeutralarXiv – CS AI · May 126/10
🧠Researchers introduce mHC-SSM, a novel architecture combining Manifold-Constrained Hyper-Connections with state space language models using stream-specialized adapters. The approach achieves significant perplexity improvements (572.91 to 461.88) on WikiText-2 benchmarks with predictable efficiency tradeoffs in throughput and memory usage.
🏢 Meta🏢 Perplexity
AINeutralarXiv – CS AI · May 126/10
🧠Researchers formalize the concept of model continuity in sequential neural networks, finding that S4 maintains stable continuous behavior while Mamba's S6 exhibits sensitivity to input amplitude despite continuous-time origins. The study establishes empirical alignment between task continuity, model continuity, and performance, with practical implications for temporal subsampling strategies.
AINeutralarXiv – CS AI · May 126/10
🧠Researchers rigorously tested claims that Mamba state-space models can discover causal structure through prediction-only training, finding the method underperforms classical approaches like PCMCI and Granger causality. The apparent success in earlier experiments was largely attributable to sample-size confounds and non-standard intervention semantics rather than genuine architectural advantages.
AINeutralarXiv – CS AI · May 126/10
🧠Researchers introduce TIDES, a new selective state space model architecture that combines the expressivity of input-dependent models like Mamba with the native irregular time-series handling of continuous-time models like S5. By moving input-dependence to the state matrix rather than the discretization step, TIDES maintains the physical meaning of time intervals while preserving per-token expressivity, achieving state-of-the-art results on time-series benchmarks.
AINeutralarXiv – CS AI · May 116/10
🧠EmambaIR introduces a novel State Space Model architecture for event-based image reconstruction that achieves superior performance over CNNs and Vision Transformers while maintaining linear computational complexity. The framework combines sparse attention mechanisms with gated state-space modules to process event camera data efficiently across motion deblurring, deraining, and HDR enhancement tasks.
AIBullisharXiv – CS AI · Apr 206/10
🧠SSMamba introduces a self-supervised hybrid state space model designed to improve pathological image classification by addressing domain shift, local-global relationship modeling, and fine-grained feature detection. The framework outperforms 11 state-of-the-art pathological foundation models on multiple public datasets without requiring large external training datasets.
AIBullisharXiv – CS AI · Mar 166/10
🧠Researchers developed a hybrid model combining Mamba-2 state space operators with Transformer blocks for recursive reasoning, achieving a 2% improvement in pass@2 performance on ARC-AGI-1 tasks with only 6.83M parameters. The study demonstrates that Mamba-2 operators can preserve reasoning capabilities while improving solution candidate coverage in tiny neural networks.
AIBullisharXiv – CS AI · Mar 55/10
🧠Researchers have developed HealthMamba, a new AI framework that uses spatiotemporal modeling and uncertainty quantification to predict healthcare facility visits more accurately. The system achieved 6% better prediction accuracy and 3.5% improvement in uncertainty quantification compared to existing methods when tested on real-world datasets from four US states.
AIBullisharXiv – CS AI · Mar 36/108
🧠Researchers introduce Mamba-CAD, a state space model using Mamba architecture for generating complex 3D CAD models from parametric sequences. The model addresses limitations in handling longer, fine-grained industrial CAD sequences through an encoder-decoder framework paired with GANs, trained on a new dataset of 77,078 CAD models.
AINeutralarXiv – CS AI · Mar 26/1015
🧠Researchers conducted an in-depth analysis of in-context learning capabilities across different AI architectures including transformers, state-space models, and hybrid systems. The study reveals that while these models perform similarly on tasks, their internal mechanisms differ significantly, with function vectors playing key roles in self-attention and Mamba layers.
