{"slug": "still-amortized-kv-cache-compaction-in-a-single-forward-pass", "title": "Still: Amortized KV Cache Compaction in a Single Forward Pass", "summary": "Researchers introduced Still, a per-layer Perceiver model that compacts KV cache in a single forward pass, enabling efficient long-context language model deployment. On Qwen and Gemma models, Still outperformed baselines by 8–22 points on the RULER benchmark across compression ratios from 8× to 200× and context lengths up to 128k tokens. The method supports iterative application for long-horizon tasks and preserves full-context performance on summarization benchmarks.", "body_md": "# Computer Science > Machine Learning\n\n[Submitted on 5 Jun 2026]\n\n# Title:Still: Amortized KV Cache Compaction in a Single Forward Pass\n\n[View PDF](/pdf/2606.07878)\n\n[HTML (experimental)](https://arxiv.org/html/2606.07878v1)\n\nAbstract:The KV cache is the memory bottleneck of long-horizon language model deployment. Practically, a deployable compactor must be lightweight enough to call during inference, expressive enough to preserve context under constraint, and reusable across a trajectory. Existing compaction methods satisfy only part of this requirement: selection methods are lightweight but subset-bound, while synthesis methods are expressive but rely on per-context optimization. Here we introduce Still, a small per-layer Perceiver trained once against a frozen base model that produces compact keys and values in a single forward pass. On Qwen and Gemma models, Still occupies the favorable side of the speed--quality frontier across compression ratios from $8\\times$ to $200\\times$ and context lengths from $8$k to $128$k. On the long-context RULER grid, Still exceeds the strongest baseline by 8--22 points. The same compact cache also supports free-form summarization, preserving most of the full-context gain on HELMET and winning a pairwise LongBench summarization comparison against KV-Distill. Because compaction is a forward pass, Still can be applied iteratively, entering a long-horizon regime unavailable to per-context methods. We show that amortization makes long-context cache compaction tractable, and synthesis makes its compact state useful at extreme compression.\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/))# 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))# 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))\nIArxiv Recommender\n\n*(*[What is IArxiv?](https://iarxiv.org/about))# 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/still-amortized-kv-cache-compaction-in-a-single-forward-pass", "canonical_source": "https://arxiv.org/abs/2606.07878", "published_at": "2026-06-14 22:29:07+00:00", "updated_at": "2026-06-14 22:41:36.897862+00:00", "lang": "en", "topics": ["large-language-models", "machine-learning", "ai-infrastructure", "ai-research", "ai-tools"], "entities": ["Qwen", "Gemma", "RULER", "HELMET", "LongBench", "KV-Distill", "Perceiver", "Still"], "alternates": {"html": "https://wpnews.pro/news/still-amortized-kv-cache-compaction-in-a-single-forward-pass", "markdown": "https://wpnews.pro/news/still-amortized-kv-cache-compaction-in-a-single-forward-pass.md", "text": "https://wpnews.pro/news/still-amortized-kv-cache-compaction-in-a-single-forward-pass.txt", "jsonld": "https://wpnews.pro/news/still-amortized-kv-cache-compaction-in-a-single-forward-pass.jsonld"}}