{"slug": "latent-sparse-attention-towards-subquadratic-ultra-long-context-attention", "title": "Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention", "summary": "A paper submitted to arXiv on 30 Oct 2026 proposes Latent Sparse Attention (LSA), an attention method that hierarchically clusters keys in a learned latent space so queries score clusters instead of every key, keeping accuracy at ultra-long context. On GLM-5.3-FLASH (QWEN-3.8-27B) at 10M tokens, the distilled variant D-LSA speeds up one attention layer by 22.7x (72.4x) in prefill and 4.6x (42.5x) in decode over its original sparse attention (full attention) while keeping similar accuracy across tasks, and the natively trained N-LSA matches full-attention accuracy with a 1.6x per-layer training speedup at 128k tokens. The authors position LSA as a way to make prefill, decode and training cheaper for agentic workflows that depend on context windows above 1M tokens.", "body_md": "# Computer Science > Artificial Intelligence\n\n  [Submitted on 30 Oct 2026]\n\n# Title:Latent Sparse Attention: Towards Subquadratic Ultra-Long Context Attention\n\n[View PDF](http://arxiv.org/pdf/2610.42147v1)\n\n[HTML (experimental)](https://arxiv.org/html/2610.42147v1)\n\nAbstract:Agentic workflows are increasingly shifting towards long-horizon tasks, such as open-ended scientific work, where models benefit from long context windows. However, at ultra-long context (above 1M tokens), full attention is prohibitively expensive. A promising approach is sparse attention, in which a dedicated indexer selects a subset of keys for the attention. Yet current methods of indexing either still have quadratic cost in context size or lose accuracy at ultra-long context scale. We address this with Latent Sparse Attention (LSA), an efficient attention method that maintains accuracy at ultra-long context. In LSA, keys are hierarchically clustered in a learned latent space, and queries score the hierarchy of clusters instead of scoring every key, making prefill, decode and training faster at long context. We propose two variants, Distilled-LSA (D-LSA), which can be applied to pre-trained models via distillation, and Native-LSA (N-LSA), which is trained natively on the next-token loss alone. On GLM-5.3-FLASH (QWEN-3.8-27B) at 10M tokens, D-LSA speeds up one attention layer by 22.7x (72.4x) in prefill and 4.6x (42.5x) in decode over its original sparse attention (full attention), while keeping similar accuracy across a variety of tasks. N-LSA matches the accuracy of full attention with a training speedup of 1.6x per layer at 128k tokens. LSA offers practical and efficient attention for models to benefit from ultra-long context.\n    \n\n### Additional Features\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/latent-sparse-attention-towards-subquadratic-ultra-long-context-attention", "canonical_source": "http://arxiv.org/abs/2610.42147v1", "published_at": "2026-11-02 15:29:39+00:00", "updated_at": "2026-11-02 15:52:37.100808+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-research"], "entities": ["arXiv", "Latent Sparse Attention", "Distilled-LSA", "Native-LSA", "GLM-5.3-FLASH", "QWEN-3.8-27B"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/latent-sparse-attention-towards-subquadratic-ultra-long-context-attention", "markdown": "https://wpnews.pro/news/latent-sparse-attention-towards-subquadratic-ultra-long-context-attention.md", "text": "https://wpnews.pro/news/latent-sparse-attention-towards-subquadratic-ultra-long-context-attention.txt", "jsonld": "https://wpnews.pro/news/latent-sparse-attention-towards-subquadratic-ultra-long-context-attention.jsonld"}}