On the Representational Limits of Quantum-Inspired 1024-D Document Embeddings: An Experimental Evaluation Framework

TL;DR

This paper evaluates quantum-inspired 1024-D document embeddings for information retrieval, revealing their structural limitations and demonstrating that they serve better as auxiliary tools rather than standalone solutions.

Contribution

It introduces an experimental framework with diagnostic tools for quantum-inspired embeddings and assesses their effectiveness across multiple domains and retrieval strategies.

Findings

BM25 outperforms quantum-inspired embeddings as a standalone method.

Quantum-inspired embeddings show weak and unstable ranking signals.

Hybrid retrieval combining lexical and embedding signals can achieve competitive results.

Abstract

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.