按年被提到
20222025 · 2 次2026
大模型与语言模型›长上下文与测试时计算·arXiv:2212.10947 · cs.CL

Parallel Context Windows for Large Language Models

Nir Ratner, Yoav Levine, Yonatan Belinkov, Ori Ram, Inbal Magar, Omri Abend, Ehud Karpas, Amnon Shashua, Kevin Leyton-Brown, Yoav Shoham · 2022
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2025-06-28 · 最近 2025-10-12
2不同的人
2活跃月份
2讨论串
2被引(OpenAlex)
arXiv 摘要(候选层不写导读)

When applied to processing long text, Large Language Models (LLMs) are limited by their context window. Existing efforts to address this limitation involve training specialized architectures, and cannot be easily applied to off-the-shelf LLMs. We present Parallel Context Windows (PCW), a method that alleviates the context window restriction for any off-the-shelf LLM without further training. The key to the approach is to carve a long context into chunks (``windows''), restrict the attention mechanism to apply only within each window, and re-use the positional embeddings across the windows. Our main results test the PCW approach on in-context learning with models that range in size between 750 million and 178 billion parameters, and show substantial improvements for tasks with diverse input and output spaces. We show additional benefits in other settings where long context windows may be beneficial: multi-hop questions and retrieval-augmented question answering with multiple retrieved documents. Our results highlight Parallel Context Windows as a promising method for applying off-the-shelf LLMs in a range of settings that require long text sequences. We make our code publicly available at https://github.com/ai21labs/parallel-context-windows.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论112233 · 2025-06

“Dropping in late into this discussion, but is there any way to "comfortably" use multiple precomputed kv-caches with current models, in the style of this work: https://arxiv.org/abs/2212.10947 ? Meaning, I pre-parse multiple documents, and the prompt and completion attention sees all of them, but there is no attention between the documents (they are all …”

MCP in LM Studio原评论 ↗
评论Palmik · 2025-10

“The observation about the "block-diagonal patterns" in RAG isn't new and has been exploited / explored before: - https://arxiv.org/abs/2410.07590 (literally titled "Block-Attention for Efficient RAG") - https://arxiv.org/abs/2409.15355v3 - https://arxiv.org/abs/2212.10947 The REFRAG paper does not cite any of these.”

Meta Superintelligence Labs' first paper is about RAG原评论 ↗