✨VideoSSM: Autoregressive Long Video Generation with Hybrid State-Space Memory
📝 Summary:
VideoSSM proposes a hybrid state-space memory model for long video generation. It unifies autoregressive diffusion with global state-space memory and local context to achieve state-of-the-art temporal consistency and motion stability. This enables scalable, interactive minute-scale video synthesis.
🔹 Publication Date: Published on Dec 4
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.04519
• PDF: https://arxiv.org/pdf/2512.04519
==================================
For more data science resources:
✓ https://xn--r1a.website/DataScienceT
#VideoGeneration #GenerativeAI #DiffusionModels #StateSpaceModels #DeepLearning
📝 Summary:
VideoSSM proposes a hybrid state-space memory model for long video generation. It unifies autoregressive diffusion with global state-space memory and local context to achieve state-of-the-art temporal consistency and motion stability. This enables scalable, interactive minute-scale video synthesis.
🔹 Publication Date: Published on Dec 4
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2512.04519
• PDF: https://arxiv.org/pdf/2512.04519
==================================
For more data science resources:
✓ https://xn--r1a.website/DataScienceT
#VideoGeneration #GenerativeAI #DiffusionModels #StateSpaceModels #DeepLearning
✨Compiler-First State Space Duality and Portable O(1) Autoregressive Caching for Inference
📝 Summary:
Mamba-2's state space model is implemented using XLA-optimized primitives, eliminating custom kernels. This enables efficient cross-platform deployment on CPU, GPU, and TPU, realizing O1 autoregressive caching with high performance.
🔹 Publication Date: Published on Mar 10
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.09555
• PDF: https://arxiv.org/pdf/2603.09555
• Github: https://github.com/CosmoNaught/mamba2-jax
==================================
For more data science resources:
✓ https://xn--r1a.website/DataScienceT
#Mamba2 #StateSpaceModels #DeepLearning #MLInference #PerformanceOptimization
📝 Summary:
Mamba-2's state space model is implemented using XLA-optimized primitives, eliminating custom kernels. This enables efficient cross-platform deployment on CPU, GPU, and TPU, realizing O1 autoregressive caching with high performance.
🔹 Publication Date: Published on Mar 10
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2603.09555
• PDF: https://arxiv.org/pdf/2603.09555
• Github: https://github.com/CosmoNaught/mamba2-jax
==================================
For more data science resources:
✓ https://xn--r1a.website/DataScienceT
#Mamba2 #StateSpaceModels #DeepLearning #MLInference #PerformanceOptimization
✨Sessa: Selective State Space Attention
📝 Summary:
Sessa is a new decoder architecture that puts attention inside a recurrent feedback path. This allows it to model long contexts better than Transformers and state-space models, achieving power-law memory decay and flexible selective retrieval. It outperforms on long-context tasks.
🔹 Publication Date: Published on Apr 21
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.18580
• PDF: https://arxiv.org/pdf/2604.18580
• Github: https://github.com/LibratioAI/sessa
==================================
For more data science resources:
✓ https://xn--r1a.website/DataScienceT
#Sessa #DeepLearning #AttentionMechanisms #StateSpaceModels #LongContextAI
📝 Summary:
Sessa is a new decoder architecture that puts attention inside a recurrent feedback path. This allows it to model long contexts better than Transformers and state-space models, achieving power-law memory decay and flexible selective retrieval. It outperforms on long-context tasks.
🔹 Publication Date: Published on Apr 21
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2604.18580
• PDF: https://arxiv.org/pdf/2604.18580
• Github: https://github.com/LibratioAI/sessa
==================================
For more data science resources:
✓ https://xn--r1a.website/DataScienceT
#Sessa #DeepLearning #AttentionMechanisms #StateSpaceModels #LongContextAI