量子启发的文档嵌入在检索中表现不稳定，仅适合作为辅助组件。
On the Representational Limits of Quantum-Inspired 1024-D Document Embeddings: An Experimental Evaluation Framework
- 基于重叠窗口与多尺度聚合构建1024维量子启发嵌入
- 独立使用时排名信号弱且不稳，混合检索才达竞争力
- 适合研究嵌入几何局限性或作为增强信号的开发者
文本嵌入是现代信息检索与检索增强生成（RAG）的核心。尽管密集型大语言模型（LLM）主导当前实践，近期研究探索了受量子启发的替代方案，源于希尔伯特空间的几何特性及其编码更丰富语义结构的潜力。本文提出一个实验框架，基于重叠窗口与多尺度聚合构建1024维量子启发文档嵌入。该流程结合语义投影（如EigAngle）、电路启发特征映射及可选的教师-学生蒸馏，并引入指纹机制以确保可复现性与受控评估。我们设计了一套混合检索诊断工具，包括静态/动态插值（BM25与嵌入得分）、候选集合并策略，以及概念性alpha-oracle作为得分融合的理论上限。在意大利语与英语的技术、叙事、法律领域控制语料上，使用合成查询进行实验，结果显示：BM25仍是强基线；教师嵌入提供稳定语义结构；独立量子启发嵌入表现出弱且不稳定的排序信号。蒸馏效果参差不齐，部分情况改善对齐但未一致提升检索性能；而混合检索在结合词法与嵌入信号时可恢复竞争力。总体表明，量子启发嵌入在几何结构上存在局限，如距离压缩与排序不稳，其角色应定位为辅助组件而非独立检索表示。
原文摘要 · Abstract (English)
Text embeddings are central to modern information retrieval and Retrieval-Augmented Generation (RAG). While dense models derived from Large Language Models (LLMs) dominate current practice, recent work has explored quantum-inspired alternatives motivated by the geometric properties of Hilbert-like spaces and their potential to encode richer semantic structure. This paper presents an experimental framework for constructing quantum-inspired 1024-dimensional document embeddings based on overlapping windows and multi-scale aggregation. The pipeline combines semantic projections (e.g., EigAngle), circuit-inspired feature mappings, and optional teacher-student distillation, together with a fingerprinting mechanism for reproducibility and controlled evaluation. We introduce a set of diagnostic tools for hybrid retrieval, including static and dynamic interpolation between BM25 and embedding-based scores, candidate union strategies, and a conceptual alpha-oracle that provides an upper bound for score-level fusion. Experiments on controlled corpora of Italian and English documents across technical, narrative, and legal domains, using synthetic queries, show that BM25 remains a strong baseline, teacher embeddings provide stable semantic structure, and standalone quantum-inspired embeddings exhibit weak and unstable ranking signals. Distillation yields mixed effects, improving alignment in some cases but not consistently enhancing retrieval performance, while hybrid retrieval can recover competitive results when lexical and embedding-based signals are combined. Overall, the results highlight structural limitations in the geometry of quantum-inspired embeddings, including distance compression and ranking instability, and clarify their role as auxiliary components rather than standalone retrieval representations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。