arXiv subjects

Yuanhong Zhong

Publications and source records attributed to Yuanhong Zhong.

5 recordsLinked to original sources

arXiv2026

Multi-Faceted Evaluation and Mitigation of Emotion Hallucinations in MLLMs

Multimodal large language models (MLLMs) have shown strong potential in open-ended emotion understanding, yet they often generate emotion hallucinations. Evaluating such hallucinations is particularly challenging for two reasons. First, emotion understanding spans multiple cognitive facets, from multimodal perception to psychological reasoning. Second, emotional interpretations are expressed in free-form language, making existing closed-ended protocols insufficient for evaluation. To address these challenges, we introduce EHR (Emotion Hallucination Rate), an evaluator that quantifies emotion hallucinations across six facets: expression, action, audio, instinct, logic, and conclusion. Using EHR, we reveal that existing mitigation methods often reduce hallucinations in some facets while aggravating them in others, exposing the limitation of coarse-grained correction and the need for facet-aware localization and mitigation. Motivated by this finding, we propose HMER (Hallucination-aware Memory-guided Emotion Reasoning), a training-free framework for emotion hallucination mitigation. HMER maintains a Hallucination Memory that records localized hallucinated claims and enables targeted logit rectification, together with an Anchor Memory that preserves reliable intermediate reasoning states to stabilize subsequent generation. By selectively suppressing unreliable cues while preserving trustworthy reasoning context, HMER enables fine-grained mitigation across diverse hallucination facets. Extensive experiments on 19 MLLMs demonstrate the prevalence of emotion hallucinations and the effectiveness of our framework across diverse model architectures.

cs.MM↗

arXiv2026

Identifiability of $g$ mode Resonances in Eccentric Binary Neutron Stars with Multidetector Observations

$g$ mode resonances in eccentric binary neutron star systems are potential probes of internal stratification, superfluidity, composition gradients, and the equation of state. Although such weak dynamical tidal signatures are unlikely to be resolved with current detector sensitivities, third generation observations may make them accessible, in which case identifying the weak resonant phase shift would provide information beyond the bulk adiabatic tidal deformability. We build a four class dataset in an eccentric harmonic framework, containing point particle, adiabatic tide, resonant $g$ mode, and pure noise samples, and use Einstein Telescope (ET) and Cosmic Explorer (CE) detector data to test whether this weak resonant phase signature can be identified from noisy time domain strain. The ET, CE, and ET+CE deep learning models reach accuracies of $0.655$, $0.815$, and $0.897$, respectively. On the same simulated samples, the matched filtering method reaches lower accuracies of $0.514$, $0.677$, and $0.689$. This result arises from the fact that the resonant correction manifests as a weak phase morphology difference superimposed on the adiabatic tidal background, whereas matched filtering is sensitive only to the overall similarity. Hence, in the presence of weak phase differences, the neural classifier employed in deep learning is better able to learn these local phase and morphology features from the complete time domain strain segment. The results indicate that joint third generation observations improve the identifiability of weak internal mode phase information.

astro-ph.HE↗

arXiv2026

Detection of Multiband Lensed Gravitational Waves from Dark Matter Halos with Deep Learning

Lensed gravitational waves acquire amplitude and phase modulations when propagating through the gravitational potential of dark matter halos, producing interference structures in the observed waveform. However, these features are often difficult to identify in detector noise. In this work, we develop a deep learning framework for the automatic classification of lensed gravitational wave signals under multiband observations. We simulate binary neutron star signals observed by the space based detector DECIGO and the ground based Einstein Telescope, and construct five classes of data including pure noise, unlensed signals, and three lensed cases generated by the SIS, CIS, and NFW dark matter halo models. By comparing single detector and joint detector configurations, we evaluate the classification performance under different observational settings. The results show that multiband observations significantly improve the identification of lensed signals and reduce confusion among different lens models. This approach provides an efficient method for automated recognition of lensed gravitational waves in future multiband gravitational wave observations.

astro-ph.IM↗

arXiv2020

Machine Learning for Nanohertz Gravitational Wave Detection and Parameter Estimation with Pulsar Timing Array

Studies have shown that the use of pulsar timing arrays (PTAs) is among the approaches with the highest potential to detect very low-frequency gravitational waves in the near future. Although the capture of gravitational waves (GWs) by PTAs has not been reported yet, many related theoretical studies and some meaningful detection limits have been reported. In this study, we focused on the nanohertz GWs from individual supermassive binary black holes. Given specific pulsars (PSR J1909$-$3744, PSR J1713$+$0747, PSR J0437$-$4715), the corresponding GW$-$induced timing residuals in PTAs with Gaussian white noise can be simulated. Further, we report the classification of the simulated PTA data and parameter estimation for potential GW sources using machine learning based on neural networks. As a classifier, the convolutional neural network shows high accuracy when the combined signal to noise ratio $\geq$1.33 for our simulated data. Further, we applied a recurrent neural network to estimate the chirp mass ($\mathcal{M}$) of the source and luminosity distance ($\text{D}_{p}$) of the pulsars and Bayesian neural networks (BNNs) to obtain the uncertainties of chirp mass estimation. Knowledge of the uncertainties is crucial to astrophysical observation. In our case, the mean relative error of chirp mass estimation is less than $13.6\%$. Although these results are achieved for simulated PTA data, we believe that they will be important for realizing intelligent processing in PTA data analysis.

astro-ph.IM↗

arXiv2019

Applying deep neural networks to the detection and space parameter estimation of compact binary coalescence with a network of gravitational wave detectors

In this paper, we study an application of deep learning to the advanced LIGO and advanced Virgo coincident detection of gravitational waves (GWs) from compact binary star mergers. This deep learning method is an extension of the Deep Filtering method used by George and Huerta (2017) for multi-inputs of network detectors. Simulated coincident time series data sets in advanced LIGO and advanced Virgo detectors are analyzed for estimating source luminosity distance and sky location. As a classifier, our deep neural network (DNN) can effectively recognize the presence of GW signals when the optimal signal-to-noise ratio (SNR) of network detectors $\geq$ 9. As a predictor, it can also effectively estimate the corresponding source space parameters, including the luminosity distance $D$, right ascension $α$, and declination $δ$ of the compact binary star mergers. When the SNR of the network detectors is greater than 8, their relative errors are all less than 23%. Our results demonstrate that Deep Filtering can process coincident GW time series inputs and perform effective classification and multiple space parameter estimation. Furthermore, we compare the results obtained from one, two, and three network detectors; these results reveal that a larger number of network detectors results in a better source location.

astro-ph.IM↗