Machine Learning Confirms GW231123 is a "Lite" Intermediate Mass Black Hole Merger

The LIGO–Virgo–KAGRA Collaboration recently reported GW231123, a binary black hole merger with a total mass of ≈190–265 M⊙, adding to the growing evidence for “lite” intermediate-mass black holes (IMBHs) with post-merger masses exceeding 100 M⊙. This event presented significant data-analysis challenges due to waveform-model systematics and coincident instrumental glitches. We present the first comprehensive machine learning–based validation of GW231123, combining three complementary frameworks: GW-Whisper, an adaptation of OpenAI’s audio transformer for gravitational-wave detection; ArchGEM, a Gaussian mixture–based density estimator; and AWaRe, a convolutional autoencoder for waveform reconstruction. Our analysis confidently identifies the astrophysical signal (>70% confidence in both detectors), characterizes the associated scattered-light glitch with physically interpretable parameters, and reconstructs the waveform with closer agreement to model-agnostic methods than to standard quasi-circular models—suggesting possible deviations such as orbital eccentricity or environmental effects. We further demonstrate the robustness of our approach on simulated mergers across 100–1000 M⊙, showing its potential to probe the IMBH regime and complement traditional LIGO–Virgo pipelines for rapid, glitch-resilient event analysis.