CLaRa：通过连续潜在推理弥合检索与生成
CLaRa: Bridging Retrieval and Generation with Continuous Latent Reasoning
AI总结：
CLaRa通过连续潜在推理框架实现检索与生成的统一优化，提升压缩和重排性能，优于现有基线方法。
摘要：
检索增强生成（RAG）通过外部知识增强大语言模型（LLMs），但仍然面临长上下文和检索生成优化不一致的问题。在本工作中，我们提出了CLaRa（连续潜在推理），一个统一的框架，该框架在共享的连续空间中执行基于嵌入的压缩和联合优化。为了获得语义丰富且可检索的压缩向量，从而减少输入生成器的文档长度，我们引入了SCP，一种基于问答和同义词监督的关键保留数据合成框架。CLaRa通过单个语言建模损失对重排器和生成器进行端到端训练，使用可微的top-k估计器使梯度通过两个模块流动。理论上，这种统一的优化将检索相关性与回答质量对齐。在多个问答基准测试中，实验表明CLaRa在压缩和重排性能上达到最先进的水平，即使在文本压缩率为16的情况下，也优于基于文本的微调基线。
英文摘要：
Retrieval-augmented generation (RAG) enhances large language models (LLMs) with external knowledge but still suffers from long contexts and disjoint retrieval-generation optimization. In this work, we propose CLaRa (Continuous Latent Reasoning), a unified framework that performs embedding-based compression and joint optimization in a shared continuous space. To obtain semantically rich and retrievable compressed vectors, thereby reducing the document length fed into the generator, we introduce SCP, a key-preserving data synthesis framework based on question answering and paraphrase supervision. CLaRa then trains the reranker and generator end-to-end via a single language modeling loss, with gradients flowing through both modules using a differentiable top-k estimator. Theoretically, this unified optimization aligns retrieval relevance with answer quality. Experiments across multiple QA benchmarks show that CLaRa achieves state-of-the-art compression and reranking performance, even at a text compression rate of 16, outperforming text-based fine-tuned baselines.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
