arXiv subjects

Andrea Volpini

Publications and source records attributed to Andrea Volpini.

6 recordsLinked to original sources

arXiv2026

Enhancing broadband second harmonic generation in a thin film lithium niobate racetrack resonator with tunable-coupling

Second harmonic generation in thin film periodically poled lithium niobate (PPLN) is constrained by an efficiency-bandwidth trade-off and fabrication-sensitive scaling. We demonstrate a racetrack resonator incorporating a short PPLN section and a tunable Mach-Zehnder interferometer coupler that enables in situ control of the coupling condition, compensating fabrication tolerances and stabilizing operation near critical coupling. The telecom pump is resonantly enhanced, while the near-infrared second harmonic is generated in single pass, eliminating dual-resonance requirements. The device achieves a 35 times efficiency enhancement over a non-resonant structure while maintaining a 7 nm bandwidth. This architecture provides a robust platform for broadband integrated frequency doubling.

physics.optics↗

arXiv2026

RLM-on-KG: Heuristics First, LLMs When Needed: Adaptive Retrieval Control over Mention Graphs for Scattered Evidence

When does an LLM controller outperform rule-based traversal for knowledge graph exploration? We study this question through RLM-on-KG, a retrieval system that treats an LLM as an autonomous navigator over an RDF-encoded mention graph for grounded question answering. Unlike GraphRAG pipelines that rely on offline LLM indexing, RLM-on-KG performs entity-first, multi-hop exploration at query time using deterministic graph construction and a fixed tool set. Our central finding is a conditional advantage: the value of LLM control depends on evidence scatter and tool-calling sophistication. The paper's core claim is LLM control versus heuristic traversal, not a generic win over GraphRAG. On GraphRAG-Bench Novel (519 questions), Gemini 2.0 Flash achieves +2.47 pp F1 over a rule-based heuristic baseline (p < 0.0001), but only +0.16 pp over a GraphRAG-local variant (not significant). With a stronger controller, Claude Haiku 4.5, the gain over heuristic grows to +4.37 pp (p < 0.001) and extends to a +2.42 pp significant improvement over GraphRAG-local (p < 0.001). The gain is largest when gold evidence is scattered across 6-10 chunks (+3.21 pp) and smallest for concentrated evidence (+1.85 pp). Cross-scale validation on MuSiQue confirms that the LLM-over-heuristic advantage transfers, with expected attenuation on smaller per-question graphs. The core architectural insight is the separation of candidate discovery from ranking: the LLM adds value through exploration breadth, while final evidence selection is best handled by pure vector re-ranking. Beyond retrieval, exploration traces provide a proposed stress-test harness for structured data quality, yielding diagnostics for coverage, connectivity, provenance, and queryability.

cs.IR↗

arXiv2026

Structured Linked Data as a Memory Layer for Agent-Orchestrated Retrieval

Retrieval-Augmented Generation (RAG) systems typically treat documents as flat text, ignoring the structured metadata and linked relationships that knowledge graphs provide. In this paper, we investigate whether structured linked data, specifically Schema.org markup and dereferenceable entity pages served by a Linked Data Platform, can improve retrieval accuracy and answer quality in both standard and agentic RAG systems. We conduct a controlled experiment across four domains (editorial, legal, travel, e-commerce) using Vertex AI Vector Search 2.0 for retrieval and the Google Agent Development Kit (ADK) for agentic reasoning. Our experimental design tests seven conditions: three document representations (plain HTML, HTML with JSON-LD, and an enhanced agentic-optimized entity page) crossed with two retrieval modes (standard RAG and agentic RAG with multi-hop link traversal), plus an Enhanced+ condition that adds rich navigational affordances and entity interlinking. Our results reveal that while JSON-LD markup alone provides only modest improvements, our enhanced entity page format, incorporating llms.txt-style agent instructions, breadcrumbs, and neural search capabilities, achieves substantial gains: +29.6% accuracy improvement for standard RAG and +29.8% for the full agentic pipeline. The Enhanced+ variant, with richer navigational affordances, achieves the highest absolute scores (accuracy: 4.85/5, completeness: 4.55/5), though the incremental gain over the base enhanced format is not statistically significant. We release our dataset, evaluation framework, and enhanced entity page templates to support reproducibility.

cs.IR↗

arXiv2025

High-purity frequency-degenerate photon pair generation via cascaded SFG/SPDC in thin film lithium niobate

Frequency-degenerate photon pairs generated using nonlinear photonic integrated devices are a crucial resource for scalable quantum information processing and metrology. However, their realization is hindered by unwanted parametric processes occurring within the same phase matching band, which degrade the signal-to-noise ratio and reduce the purity of the associated quantum states. Here, we propose a dual-pump scheme to produce frequency-degenerate photon pairs, based on cascaded sum-frequency generation and spontaneous parametric down-conversion occurring within a single waveguide, while strongly suppressing parasitic photon pair generation from single-pump processes. This approach significantly simplifies the design compared to microresonator-based methods and enables both pumping and collection of photon pairs entirely in the telecom band. We experimentally validate the concept in a layer-poled thin film lithium niobate waveguide, achieving frequency-degenerate photon pair generation with a brightness of \SI{1.0(3)e5}{\hertz \per \nm \per \square \milli \watt } and a 40 dB suppression of unwanted single-pump processes.

quant-ph↗

arXiv2023

Monolithically Integrated Wavelength-meter in InP with measurement bandwidth of 100nm centered on the C band

In this paper we will explore the creation of a monolithically integrated wavelength meter in InP. This type of devices are a key requirement for many applications and it is especially important to have them integrated with active components like lasers and gain sections. We present a wavelength meter based on multiple ring resonators that has been realized in a commercial MPW run and tested using a tunable laser. The designed circuit is theoretically capable of resolution down to 1.6pm and a measurement speed down to 500ps within a wavelength range of 100nm.

physics.optics↗

arXiv2019

Time response of a microring resonator to a rectangular pulse in different coupling regimes

We discuss the analytical temporal response of a microring resonator excited through a bus waveguide by an optical rectangular pulse. Finite difference time domain (FDTD) simulations illustrate the analytical solution and help in understanding the meaning of the different coupling regimes. In addition, we show that the temporal dynamics allows determing the coupling regime while the commonly used spectral characterization in the stationary regime does not. We also take advantage of the simulation to highlight the phase shift between the input and the output signals in the different coupling regimes. Finally, measurements on a Si3N4 microring resonator are performed and analyzed in the case of under-coupling regime to illustrate how the time response study leads to the Q-factor determination.

physics.optics↗