AI 服務暫時不可用，以下為來源正文，待恢復後補全翻譯。
[Submitted on 16 Sep 2026] Title:Elastic Threshold Attention: Learned Contextual Sparsity for Long-Context Decoding View a PDF of the paper titled Elastic Threshold Attention: Learned Contextual Sparsity for Long-Context Decoding, by Themistoklis Haris and 1 other authors View PDF HTML (experimental) Abstract:Massive KV caches can cause severe memory-bandwidth bottlenecks during long-context decoding. Sparse attention methods mitigate this via selective loading, but that comes at a cost: rigid heuristics drop necessary context, leading to quality degradation. We introduce \textbf{Elastic Threshold Attention (ETA)}, an end-to-end trainable architecture that achieves hardware-accelerated decoding speed without sacrificing dense model quality. ETA predicts dynamic, contextual thresholds directly from query representations, allowing the model to allocate dense-like context to difficult retrieval or reasoning steps while pruning routine tokens. To learn this policy from scratch without representation collapse, ETA \emph{multiplicatively suppresses} sub-threshold logits toward zero during training rather than deleting them. Training against this smooth uniform attention floor provides a distributed probability reservoir that \textbf{causes localized attention sinks on initial tokens to disappear}. It also enables the model to hard-prune uninformative KV blocks at inference time and absorb incidental tokens co-admitted by coarse GPU block selection. As a result, a 1.45B pretrained ETA model rivals dense attention across language modeling, commonsense reasoning, and long-context needle retrieval at $\approx 85\%$ training sparsity and $\approx 38\%$ active decode density. At inference time, we implement a custom decode kernel in Triton that screens KV blocks in $O(1)$ time using cached geometric-probabilistic bounds, delivering up to $2.5\times$ wall-clock decode speedups over FlashAttention-2 on sequences up to 512K tokens. Finally, we introduce an offline calibration algorithm for domain-specific deployments that freezes per-head constant thresholds to eliminate predictor overhead, cutting attention compute by an additional $27\%$. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2609.20888 [cs.LG] (or arXiv:2609.20888v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2609.20888 arXiv-issued DOI via DataCite (pending registration) Submission history From: Themistoklis Haris [view email] [v1] Wed, 16 Sep 2026 23:23:10 UTC (530 KB) Full-text links: Access Paper: View a PDF of the paper titled Elastic Threshold Attention: Learned Contextual Sparsity for Long-Context Decoding, by Themistoklis Haris and 1 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.LG new | recent | 2026-09 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Loading... Data provided by: Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both 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. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)