{"version":"1.0","type":"card","id":"651ccf46-5cd6-436c-a259-81ca8418285a","url":"https://stacklist.com/card/651ccf46-5cd6-436c-a259-81ca8418285a","title":"Parallel Context Windows for LLMs — arXiv Seminal Paper","source_url":"https://arxiv.org/abs/2212.10947","note":"The research paper that introduced parallel context processing as a way to overcome fixed-length limits. A foundational read if you want to understand how modern long-context architectures evolved — and why some models handle multi-document tasks so well.","image":{"url":"https://ucarecdn.com/5687905b-7371-4d7b-ada6-1b0e2f6a83c2/","alt":"Parallel Context Windows for LLMs — arXiv Seminal Paper","width":336,"height":96},"stack":{"id":"e16dcdcb-06b9-481c-aef9-40b6d73a5c8e","title":"Master AI & LLM Context: From Basics to Production","url":"https://stacklist.com/c/technology/stack/e16dcdcb-06b9-481c-aef9-40b6d73a5c8e"},"created_at":"2026-07-02T10:09:06.527Z","updated_at":null,"aco":{"summary":"Parallel Context Windows (PCW) is a proposed method that extends the effective context window of off-the-shelf large language models without additional training by splitting long contexts into chunks, restricting attention within each window, and reusing positional embeddings. The paper demonstrates substantial improvements on in-context learning, multi-hop questions, and retrieval-augmented QA tasks across models ranging from 750 million to 178 billion parameters, and was presented at ACL 2023.","tags":["parallel-context-windows","large-language-models","context-window","in-context-learning","attention-mechanism","positional-embeddings","natural-language-processing"],"key_entities":[{"name":"Parallel Context Windows (PCW)","type":"concept","confidence":0.99},{"name":"Nir Ratner","type":"person","confidence":0.95},{"name":"Yoav Levine","type":"person","confidence":0.95},{"name":"Yonatan Belinkov","type":"person","confidence":0.9},{"name":"Ori Ram","type":"person","confidence":0.9},{"name":"Amnon Shashua","type":"person","confidence":0.9},{"name":"Kevin Leyton-Brown","type":"person","confidence":0.9},{"name":"Yoav Shoham","type":"person","confidence":0.9},{"name":"Large Language Models","type":"technology","confidence":0.98},{"name":"ACL 2023","type":"event","confidence":0.97},{"name":"arXiv","type":"organization","confidence":0.95},{"name":"in-context learning","type":"concept","confidence":0.92},{"name":"retrieval-augmented question answering","type":"concept","confidence":0.88},{"name":"attention mechanism","type":"concept","confidence":0.9}],"classification":"reference","language":"en","confidence":0.85,"provenance":{"model":"claude-opus-4-6","tool":"@stacklist/mcp-server@2.0.0","confidence":0.85,"timestamp":"2026-07-02T10:09:17.699Z"},"token_counts":{"approximate":1105,"cl100k":1054},"content_hash":"sha256:4d88db2e55e22eb12c863924156e9040e7371425635ca64a1bb3738016cea164","acp_version":"0.2","body_available":true,"body_tokens":1105,"visibility":"public","agent_accessible":true,"status":"final"},"_links":{"self":"/api/public/card/651ccf46-5cd6-436c-a259-81ca8418285a.json","html":"https://stacklist.com/card/651ccf46-5cd6-436c-a259-81ca8418285a","md":"/api/public/card/651ccf46-5cd6-436c-a259-81ca8418285a.md","stack_json":"/api/public/stack/e16dcdcb-06b9-481c-aef9-40b6d73a5c8e.json"}}