arXiv subjects

Tzu-Hsuan Lin

Publications and source records attributed to Tzu-Hsuan Lin.

5 recordsLinked to original sources

arXiv2026

FROAV: A Framework for RAG Observation and Agent Verification -- Lowering the Barrier to LLM Agent Research

The rapid advancement of Large Language Models (LLMs) and their integration into autonomous agent systems has created unprecedented opportunities for document analysis, decision support, and knowledge retrieval. However, the complexity of developing, evaluating, and iterating on LLM-based agent workflows presents significant barriers to researchers, particularly those without extensive software engineering expertise. We present FROAV (Framework for RAG Observation and Agent Verification), an open-source research platform that democratizes LLM agent research by providing a plug-and-play architecture combining visual workflow orchestration, a comprehensive evaluation framework, and extensible Python integration. FROAV implements a multi-stage Retrieval-Augmented Generation (RAG) pipeline coupled with a rigorous "LLM-as-a-Judge" evaluation system, all accessible through intuitive graphical interfaces. Our framework integrates n8n for no-code workflow design, PostgreSQL for granular data management, FastAPI for flexible backend logic, and Streamlit for human-in-the-loop interaction. Through this integrated ecosystem, researchers can rapidly prototype RAG strategies, conduct prompt engineering experiments, validate agent performance against human judgments, and collect structured feedback-all without writing infrastructure code. We demonstrate the framework's utility through its application to financial document analysis, while emphasizing its material-agnostic architecture that adapts to any domain requiring semantic analysis. FROAV represents a significant step toward making LLM agent research accessible to a broader scientific community, enabling researchers to focus on hypothesis testing and algorithmic innovation rather than system integration challenges.

cs.LG↗

arXiv2025

Investigating the Anti-Correlation between Photon Index and Flux of the Crab using RXTE and NuSTAR

We present a systematic study of the recently reported anti-correlation between X-ray flux and photon index ($\Gamma$) in the Crab Nebula, using archival RXTE/PCA (3 - 50 keV), RXTE/HEXTE (20 - 100 keV), and NuSTAR (3 - 78 keV) observations. Spectra were extracted in soft (3 - 10 keV) and hard bands (10 - 50 keV, 10 - 78 keV, 20 - 100 keV) and fitted with an absorbed power-law model. Across all instruments and energy ranges, we confirm the existence of a persistent negative correlation -- harder spectra at higher flux levels. The correlation is stronger in the hard bands compared to the soft bands. This is consistent with synchrotron emission modulated by magnetic field variations in the pulsar wind nebula.

astro-ph.HE↗

arXiv2024

Quantum Information-Empowered Graph Neural Network for Hyperspectral Change Detection

Change detection (CD) is a critical remote sensing technique for identifying changes in the Earth's surface over time. The outstanding substance identifiability of hyperspectral images (HSIs) has significantly enhanced the detection accuracy, making hyperspectral change detection (HCD) an essential technology. The detection accuracy can be further upgraded by leveraging the graph structure of HSIs, motivating us to adopt the graph neural networks (GNNs) in solving HCD. For the first time, this work introduces quantum deep network (QUEEN) into HCD. Unlike GNN and CNN, both extracting the affine-computing features, QUEEN provides fundamentally different unitary-computing features. We demonstrate that through the unitary feature extraction procedure, QUEEN provides radically new information for deciding whether there is a change or not. Hierarchically, a graph feature learning (GFL) module exploits the graph structure of the bitemporal HSIs at the superpixel level, while a quantum feature learning (QFL) module learns the quantum features at the pixel level, as a complementary to GFL by preserving pixel-level detailed spatial information not retained in the superpixels. In the final classification stage, a quantum classifier is designed to cooperate with a traditional fully connected classifier. The superior HCD performance of the proposed QUEEN-empowered GNN (i.e., QUEEN-G) will be experimentally demonstrated on real hyperspectral datasets.

cs.CV↗

arXiv2024

Energy calibration of GTM on ground

The Gamma-ray Transients Monitor (GTM) on board the Formosat-8B (FS-8B) satellite is designed to detect and localize Gamma-Ray Bursts (GRBs). By utilizing 2+2 CITIROC chips to manipulate 4+4 detectors, which are composed of GAGG(Ce) scintillators coupled with Silicon Photomultipliers (SiPMs) and oriented in various directions to achieve all-sky coverage, the GRB saturation fluences of GTM in the 50 keV to 1 MeV range for Short GRBs (SGRBs) and Long GRBs (LGRBs) were estimated to be about $3.1 \times 10^{-4}$ and $5.0 \times 10^{-3}\ {\rm erg/cm^2}$, respectively, based on simulations. To precisely interpret the GTM readout signal in terms of energy, several measurements for isotope and gain calibration were conducted. Despite encountering issues with crosstalk and SiPM saturation effect in the data, the energy spectrum can still be recovered by appropriately discarding channel noise and mapping with the correct ADC-to-energy relation. This paper summarizes the energy resolution of GTM and the linear variations in the relationship between photon energy and readout signal. At 662 keV, the energy resolution is about 16 %. Also, it demonstrates that greater gain is achieved by increasing voltage or decreasing temperature.

astro-ph.IM↗

arXiv2023

Observational connection of non-thermal X-ray emission from pulsars with their timing properties and thermal emission

The origin and radiation mechanisms of high energy emissions from pulsars have remained mysterious since their discovery. Here we report, based on a sample of 68 pulsars, observational connection of non-thermal X-ray emissions from pulsars with their timing properties and thermal emissions, which may provide some constraints on theoretical modeling. Besides strong correlations with the spin-down power $\dot{E}$ and the magnetic field strength at the light cylinder $B_{\rm lc}$, the non-thermal X-ray luminosity in 0.5 - 8 keV, $L_{\rm p}$, represented by the power-law component in the spectral model, is found to be strongly correlated with the highest possible electric field strength in the polar gap, $E_{\rm pc}$, of the pulsar. The spectral power index $Γ_{\rm p}$ of that power-law component is also found, for the first time in the literature, to strongly correlate with $\dot{E}$, $B_{\rm lc}$ and $E_{\rm pc}$, thanks to the large sample. In addition, we found that $L_{\rm p}$ can be well described by $L_{\rm p}\propto T^{5.96\pm 0.64}R^{2.24\pm 0.18}$, where $T$ and $R$ are the surface temperature and the emitting-region radius of the surface thermal emission, represented by the black-body component in the spectral model. $Γ_{\rm p}$, on the other hand, can be well described only when timing variables are included, and the relation is $Γ_{\rm p} = \log(T^{-5.8\pm 1.93}R^{-2.29\pm 0.85}P^{-1.19\pm 0.88}\dot{P}^{0.94\pm 0.44})$ plus a constant. These relations strongly suggest the existence of connections between surface thermal emission and electron-positron pair production in pulsar magnetospheres.

astro-ph.HE↗