arXiv subjects

Luca Cirfeta

Publications and source records attributed to Luca Cirfeta.

7 recordsLinked to original sources

arXiv2026

A Content-Addressed Workflow for Reproducible DANTE Gravitational-Wave Anomaly Analysis

Unsupervised analysis of gravitational-wave detector data combines expensive scientific stages with data acquisition, provenance checks, statistical calibration, follow-up products, and reporting. Reproducing such an analysis requires more than preserving model weights or a final candidate table: the execution graph, scientific contracts, retry semantics, and verification boundary must also be explicit. We present a local workflow layer for the Domain-Adaptive Network for Transient Evaluation (DANTE) that represents a corrected O4a analysis as a content-addressed, 15-stage directed acyclic graph. The architecture separates immutable stage evidence from mutable operational progress, binds command-line and browser interfaces to the same run identity, and resumes only work compatible with the frozen contract. The tagged v1 release records all 15 stages as verified in a machine-readable receipt and passed a documented human usability gate. The release verifies adopted scientific artifacts rather than recomputing them, and therefore makes no new claim of global significance, astrophysical discovery, or public real-time operation. This paper describes the architecture, its fail-closed controls, and the bounded evidence supporting reproducible local use.

astro-ph.IM↗

arXiv2026

Stress-Testing DANTE under Detector Domain Shift: a Representation-Coherent Reanalysis of LIGO O4a

This sixth version of the Domain-Adaptive Network for Transient Evaluation (DANTE) preprint stress-tests an unsupervised transient-noise pipeline under representation mismatch and observing-run adaptation. We reanalyse 10,429 detector-time strain candidates from 42 LIGO O4a sessions using frozen DINOv2 patch embeddings and a Top-k multiple-instance score. Candidate and native-background Q-transforms share Q in [4,64], and detector-specific thresholds are calibrated from 5,000 run-native windows by temporal-block bootstrap. The coherent analysis yields 6,365 ROBUST, 1,275 AMBIGUOUS, and 2,789 BACKGROUND statistical dispositions; 4,676 of 10,372 paired historical dispositions differ from the cross-representation v5 analysis. Direct controls resolve an O3b-O4a score shift and reduction after native adaptation for H1, but not L1, while known-glitch separation is detector- and morphology-dependent. Replicated studies quantify population-dependent sensitivity to background draw, clustering seed, dictionary size, and whitening, demonstrate native-index absorption, and identify conditional low-Q blindness. A conservative H1-L1 max-shift screen yields 13/8,806 values above threshold, but its on-source values and pooled per-event null maxima are not exchangeable. The primary two-null PEM endpoint shows no resolved ROBUST-BACKGROUND enrichment (p=1.000), and two catalogue overlaps are consistent with a circular-shift coverage proxy (p=0.651). Simulation-only compact-binary controls show detector- and distance-dependent disagreement between novelty, native disposition, and physical coincidence. We withdraw the v5 discovery, rate-limit, catalogue-recall, and survey-wide stability interpretations. The supported result is a measured set of failure modes and validity conditions for unsupervised detector characterization, not a new glitch class or an astrophysical search.

astro-ph.IM↗

arXiv2026

An Unsupervised Search for Novel Instrumental Glitches in LIGO O4a: Multi-Scale Sensitization, Empirical Physical Vetoes, and Rate Upper Limits

The fourth observing run (O4) of Advanced LIGO, Virgo, and KAGRA presents unparalleled sensitivity, rendering unsupervised pipelines highly vulnerable to the non-stationary domain shift of the detectors' noise manifolds. We present DANTE V3, concluding a longitudinal investigation into unmodeled anomalies during early O4a. By expanding to a multi-scale geometric framework (0.5 s to 4.0 s), we amplify morphological sensitivity, extracting 10,372 unique candidates. To mitigate domain-shift artifacts, we introduce a block-bootstrap Domain Shift Defense (DSD) against a vector-quantized native background index. While 28.3% of candidates survive recalibration, global topological analysis reveals they lack discrete morphological cohesion. The survivors coalesce into a single macro-cluster without compact substructure; we show explicitly that morphological "diffusivity" comparisons used previously are confounded and retire them. Pristine background is even more monolithic (100% in one cluster vs 99.77% for survivors): this topology is a property of the embedding geometry, not of anomaly status. Executing a definitive physical environment monitoring (PEM) cross-correlation defense using an empirically calibrated null, one singleton is vetoed by a control-line coupling, while another survives as an uncatalogued instrumental outlier. Replacing the embedding-similarity cross-detector coincidence test with a physical normalized cross-correlation test, we find no coincident events among 8,749 candidates. We quote 90% frequentist Poisson upper limits on the rate of novel uncatalogued instrumental morphologies --- $R_{90} \le 5.83 \mathrm{yr}^{-1}$ (H1) and $R_{90} \le 5.63 \mathrm{yr}^{-1}$ (L1). This underscores the absolute necessity of native background recalibration and physical auxiliary vetoes in unsupervised gravitational-wave astronomy.

astro-ph.IM↗

arXiv2026

DANTE: A Reference-Guided Unsupervised Pipeline for Extended-Transient Anomaly Characterization in LIGO O4a

The analysis of gravitational-wave detector data during the fourth observing run (O4) requires robust methods to distinguish stationary instrumental noise from non-stationary transients (glitches). In this work, we present DANTE (Domain-Adaptive Network for Transient Evaluation), a pipeline designed to discover and triage novel non-stationary artifacts entirely without labels. We demonstrate that adapting a pre-trained Vision Transformer (DINOv2) to extract local patch embeddings from time-frequency spectrograms allows for high-resolution mapping of transient anomalies. We formalize the Signal Dilution Barrier via controlled injection tests, showing that while Multiple Instance Learning (MIL) Top-k pooling recovers extended topologies, it is blind to sub-second morphologies. To address small-sample taxonomy instability, we introduce an adaptive Dirichlet Process Mixture Model (DPMM) that dynamically selects covariance structures. Finally, by implementing a native O4a background recalibration, we resolve the domain-shift problem, demonstrating consistency with the hypothesis that pervasive O4a morphologies (initially flagged as novel by historical references) are stationary artifacts. We conclude that unsupervised anomaly detection strictly requires native recalibration to filter domain-shift artifacts, while definitive classification of remaining unmodeled singletons requires multi-channel validation.

astro-ph.IM↗

arXiv2026

Patch-Level DINOv2 Scoring for Gravitational-Wave Glitch Detection: Breaking the Signal Dilution Barrier via Vector-Quantized Local Feature Indexing

We present a patch-level scoring architecture for unsupervised gravitational-wave glitch detection that mitigates the signal dilution limitation identified in Cirfeta (2026b). The CLS token of frozen DINOv2 (ViT-S/14) performs global average pooling over 37x37=1369 patches, systematically suppressing signals occupying less than 5% of the spectrogram grid. We replace the global CLS similarity metric with a top-$k$ order statistic over individual patch token similarities against a Vector-Quantized reference index ($K=64$ centroids per class, 19 Gravity Spy O3b morphologies, 1216 total centroids). Applied to strain-domain injections in LIGO O4a L1 data (session 20260524), we demonstrate a statistically significant distributional separation ($\text{KS}=0.963$ at optimal $k=68$) for spatially extended morphologies (SpiralBurst), while confirming the patch-size temporal resolution limit for ultra-short transients (AsymBlip). A topological saliency map constructed from spatial patch similarity against a background matrix (78 null segments) correctly localizes glitch signatures for Scattered_Light and injected SpiralBurst. The Max/Mean ratio analysis demonstrates that patch-level saliency functions as a topological visualizer rather than a binary detector, consistent with the non-isotropic geometry of DINOv2 embedding space on GW spectrograms.

astro-ph.IM↗

arXiv2026

Sensitivity Limits and Operational Threshold Calibration for DINOv2-based Gravitational-Wave Glitch Characterization: A Strain-Domain Mock Data Challenge on LIGO O4a

We present a Mock Data Challenge (MDC) to characterize the sensitivity limits of the gravi-signal-ml pipeline (Cirfeta 2026) for unsupervised gravitational-wave glitch detection. Strain-domain synthetic injections of eight morphological families into public LIGO O4a L1 data reveal two threshold-dependent sensitivity regimes. With a session-adaptive dynamic threshold (tau_dyn = mu_bg - 2.5 * sigma_bg), the pipeline recovers visually anisotropic morphologies (Butterfly, ZSweep) at matched-filter SNR >= 70, reaching Recall = 1.0, though the False Positive Rate (FPR) remains uncontrolled across sessions. Characterization of the full O4a embedding distribution (N = 188,142 segments) reveals extreme non-Gaussianity (skewness = -4.12, excess kurtosis = 15.38, Shapiro-Wilk p-value near 0), with the left tail best modeled by a Generalized Extreme Value (GEV) distribution. Under a statistically rigorous operational threshold (tau_op = 0.874) calibrated at the empirical 5x10^-5 quantile (FPR < 0.01%), the MDC yields Recall = 0 for all eight morphologies at all tested SNR levels, including narrowband structures (HarmonicComb, NarrowChirp) and impulsive transients (AsymBlip) at SNR up to 430. We trace this insensitivity to the global average pooling of the DINOv2 [CLS] token, which dilutes signals occupying a small fraction (<5%) of the spectrogram's 37x37 patch grid. The null result of Cirfeta (2026) is conditionally reinterpreted: it confirms the absence of novel macro-structures but cannot exclude localized micro-structures. These findings provide a quantitative roadmap for next-generation ViT-based pipelines using patch-level scoring and multi-scale windowing.

astro-ph.IM↗

arXiv2026

Unsupervised Morphological Characterization of Gravitational-Wave Glitches in LIGO O4a Using Frozen DINOv2 Features

A central open question in gravitational-wave detector characterization is whether the O4a observing run has introduced glitch morphologies not present in earlier runs. We present gravi-signal-ml, an open-source pipeline for unsupervised morphological characterization of instrumental noise transients (glitches) in LIGO gravitational-wave data, applied to 1,277 hours of public O4a strain data from the Hanford and Livingston detectors. The pipeline extracts 384-dimensional visual embeddings from Q-transform spectrograms using a frozen DINOv2 Vision Transformer with register tokens (ViTS/14), requiring no labeled training data. Embeddings are projected via PCA and UMAP with cosine metric, then clustered using a Dirichlet Process Mixture Model (DPMM). Cluster robustness is systematically assessed through ablation studies, stability analysis across hyperparameter perturbations, and morphological cross-check against an in-domain Gravity Spy O3b reference index. A time-slide background test excludes statistically significant H1--L1 coincidences ($p \geq 0.1$) in all sessions. Across 188,000+ spectrograms, no morphologically novel glitch candidates were identified -- all anomalous clusters map to known Gravity Spy classes with cosine similarity $> 0.98$. L1 embeddings show consistently high robustness (ablation ARI $> 0.90$ in all four sessions), while H1 exhibits lower and more variable grayscale ablation ARI ($\sim 0.68$--$0.90$), suggesting a structural difference in the H1 noise manifold under DINOv2 feature extraction. This null result, obtained with a fully validated pipeline, establishes a reproducible baseline for zero-shot glitch morphology characterization in O4a data. The pipeline and all results are publicly available at https://github.com/lucacirfeta/dante-gravi-signal-ml DOI: https://doi.org/10.5281/zenodo.20121860.

astro-ph.IM↗