arXiv subjects

Zhaoran Liu

Publications and source records attributed to Zhaoran Liu.

At least 19 recordsLinked to original sources

arXiv2026

Overmassive No More: The Case for Little Red Dots Hosting Black Hole Seeds as Massive as Single Supermassive Stars

Little Red Dots (LRDs) display singular properties unlike any known class of AGN or galaxies, motivating novel mass estimators for their central engines. Inspired by their similarities to stellar phenomena, here we interpret the LRD continuum as being produced by a pseudo-photosphere. We fit tailored stellar atmosphere models to host-subtracted LRD central engines ("black hole stars," BH*s) represented by stacks of $117$ objects. Typical BH* continuum spectra are well fit by models in a narrow range of temperatures ($T_{\rm eff}\approx4200-4800$ K), with bolometric luminosities $\approx10^{43-45}$ erg s$^{-1}$, implying pseudo-photospheric radii $\approx700-2000$ au. Based on these parameters, we explore four different approaches to deriving BH* masses: 1) using the surface gravity from atmosphere models; 2) appealing to the resemblance to super-Eddington phenomena; 3) approximating the escape velocity from the outflowing material; and 4) exploiting the lack of variability to bound the dynamical time. For the typical BH*, all of these methods yield remarkably consistent masses of $\approx10^{4-5}\,M_\odot$, implying a highly super-Eddington luminosity of $L_{\rm{bol}}/L_{\rm{Edd}}\sim5-50$. These mass estimates place BH*s within the scatter of the local scaling relation between black hole mass and host galaxy stellar mass, providing a self-consistent alternative to "overmassive" black holes that lie $2-3$ dex above it. Crucially, our derived masses are consistent with BH*s arising from single supermassive stars (SMSs), whose masses cannot exceed $\approx10^{5-6}\,M_\odot$ due to general relativistic instabilities. Furthermore, for our derived $L_{\rm bol}/L_{\rm Edd}$, the sharp cutoff of the LRD luminosity function matches the maximum theoretical mass of an SMS. With LRDs, we may therefore be directly observing the birth of heavy black hole seeds.

astro-ph.GA↗

arXiv2026

Can Little Red Dots Contribute To The Early Universe Cosmic Dust Budget?

The origin of dust in the early Universe remains uncertain. We explore whether Little Red Dots (LRDs) could provide a high-redshift dust-production channel. Motivated by an analogy with Type IIn supernovae, we investigate whether LRDs may share the efficient dust-forming conditions. We consider dust formation in outer winds and later in a shielded cold dense shell after the central source fades. Such dust may avoid a global remnant-phase reverse shock, while longer LRD lifetimes may promote grain growth. We model the SEDs of two low-redshift LRD analogs as a thermal pseudo-photosphere plus optically thin dust components, obtaining dust reservoirs of order $10^{2}$-$10^{3}\,M_{\odot}$. Although photometry cannot exclude pre-existing dust, the narrow Balmer components of both analogs are close to the Case B ratio and their narrow-line environments are metal-poor, disfavoring a dominant diffuse host-ISM origin for the inferred reservoir. Formation within the LRD outflow is consistent with our optical-depth, sublimation, and energy-balance checks. In this scenario, the large dust masses do not produce a strong optical attenuation because a clumpy or asymmetric distribution can leave the dominant optical sightlines relatively unobscured. Our population calculation shows that the LRD contribution to early-Universe dust production can range from negligible to dominant. At higher efficiencies, LRDs can approach or exceed the lower CCSN contribution above $z\approx5$, whereas less favorable assumptions yield a minor contribution. We therefore propose that LRDs can act as early-Universe dust factories and, under some conditions, may dominate over CCSNe or provide seed grains for subsequent growth in the early ISM.

astro-ph.GA↗

arXiv2026

BEACON: JWST NIRCam Pure-parallel Imaging Survey. III. Constraints on the UV LF and the Clustering of z~7-14 Galaxies

The James Webb Space Telescope (JWST) has extended the frontier of galaxy detection to redshifts z>11, finding a high abundance of UV-bright sources that challenge theoretical models. However, most current results come from just a few fields, introducing uncertainties due to cosmic variance. Here, we constrain z~7-14 UV luminosity functions (LFs) over ~400 sq. arcmin across 36 independent sightlines from DR2 of BEACON, a JWST pure-parallel NIRCam multi-band imaging survey. We identify 164 7 13 do not rule out a possible excess. We measure significant clustering of bright (Muv<-20.5) galaxies at 7<z<10. Fields hosting such sources are approximately three times more likely to be overdense relative to the full survey, implying that UV-bright galaxies preferentially reside in the most massive halos at these redshifts. Comparing with semi-numerical simulations, we estimate that Muv < -20.5 galaxies inhabit halos ~0.8 dex less massive at z~11 than at z~8, consistent with a shift to higher star formation rates. However, their observed clustering exceeds predictions from pre-JWST luminosity-halo mass relations, suggesting these sources reside in more massive halos than previously modelled and/or multiple halo occupation.

astro-ph.GA↗

arXiv2026

BEACON: JWST NIRCam Pure-parallel Imaging Survey. IV. A Systematic Search for Galaxy Overdensities and Evidence for Gas Accretion Mode Transition

We systematically search for galaxy overdensities using 20 independent fields with a minimum of six filters (F090W, F115W, F150W, F277W, F356W, and F444W) from BEACON, the JWST Cycle 2 NIRCam pure-parallel imaging survey. We apply an adaptive kernel-density estimation method that incorporates the full photometric redshift probability distribution function of each galaxy to map galaxy overdensities, and identify 207 significant ($>4\,σ$) overdensities at $1.5 3$, potentially due to a larger cold gas supply in earlier times. Our study reveals a complex interplay between individual galaxies and large-scale environmental properties, marking the onset of environmental effects on galaxy quenching in massive halos at cosmic noon.

astro-ph.GA↗

arXiv2026

Little Red Dots at z~2 in EIGER reveal a gentle decline with respect to their peak number density at z~5

We report the discovery of a sample of little red dots (LRDs) at $z \approx 2$ identified from deep JWST/NIRCam imaging and wide-field slitless spectroscopy over $140$ arcmin$^2$ from the EIGER survey. With an improved blind broad-line identification algorithm, we select 19 sources at spectroscopic redshifts $z = 1.55-3.18$ identified via rest-frame near-infrared lines (Paschen-$β$, HeI+Pa$γ$ and OI). Based on a range of spectro-photometric criteria, we classify five of these sources as LRDs and the other 14 as classical active galactic nuclei (AGNs). This classification is corroborated by some X-ray detections among the AGNs. Classical AGNs dominate the number counts above optical luminosities M$_{5100}<-22.5$, whereas the LRD fraction among broad-line sources reaches 100 % at M$_{5100}\approx-20$. The LRDs span the range in Balmer break strengths seen in the higher redshift populations. Blue-shifted HeI absorption is detected in the two reddest sources. The HeI/Pa$γ$ ratio cleanly separates LRDs from classical AGNs and seems to anti-correlate with Balmer break strength, likely tracing HeI self-absorption at higher gas column densities. Our LRD sample has a similar optical luminosity range as their high-redshift counterparts, corresponding to black hole masses of $\sim10^{6}$ M$_{\odot}$ at the Eddington luminosity. We measure LRD number densities of $\approx 7\times10^{-6}$ cMpc$^{-3}$ at $z = 1.9-2.5$, which indicates that LRDs represent $\lesssim 3$ % of the AGN population at these epochs. Our results confirm the previously reported decline in the LRD number density with respect to $z \approx 5$ based on photometric surveys, although we find the decline to be more gentle than earlier emphasized.

astro-ph.GA↗

arXiv2026

Little Red Dots as Intermediate Mass, Super-Eddington Engines: Insights from Type IIn Supernovae and The 1837-1856 Great Eruption of $η$ Carinae

JWST's Little Red Dots (LRDs) display a unique constellation of features that do not occur simultaneously in any other class of galaxies or AGN. Here we observe that many of these features find parallels in the 19th century Great Eruption (GE) of $η$ Carinae and a sub-class of supernovae (Type IIn). Drawing on these stellar phenomena -- outflows trapped by dense circumstellar gas envelopes -- we sketch a possible scenario for LRDs. Outflows from the central engine produce an enshrouding envelope of gas that may be thought of as a slow wind. This dense wind and its enormous extent produce an opacity so high that a pseudo-photosphere forms within the wind, obscuring the central engine and manifesting as a blackbody-like continuum. Radiation from the buried engine powers the system. The engine may also launch fast winds that crash into the existing envelope to generate shocks. Lines form within the wind above the photosphere -- electron scattering and absorption in the clumpy (ionized + neutral) medium account for broad wings and P-Cygni cores. A key implication is that inferences of ``overmassive black holes" may be interpreting this wind-like physics as a virial broad-line region. We propose an escape velocity argument to constrain the mass of the engine, which yields $M<10^{5} M_\odot$ for the typical LRD. The lack of variability and low surface gravity of the photosphere provide further support for intermediate mass ($M\approx10^{3-6} M_\odot$), but very luminous super-Eddington ($L_{\rm{bol}}/L_{\rm{edd}}\gtrsim5$) systems harboring a supermassive star or intermediate mass black hole. Paralleling the evolution of IIn SNe, dust production in the envelope may mark the beginnings of classical AGN. This paper explores a possible self-consistent explanation for the entire life-cycle of LRDs, from their enshrouding in dense gas to their fates as seeds of massive black holes.

astro-ph.GA↗

arXiv2026

Black Hole Stars Across the Universe: Identifying Central Engine Dominated Little Red Dots at $z\sim1.5-9.5$

Photometric selections of Little Red Dots (LRDs) largely rely on identifying their ``V-shaped'' spectral energy distribution (SED). Recent work suggests this V-shape stems from a combination of a central engine -- also referred to as a Black Hole Star (BH*) -- and a star-forming host galaxy. We present a new and highly complementary photometric selection that is based on incorporating BH* templates in the \texttt{eazy} redshift fitting code. Selecting compact sources where a BH* template contributes $>80$\% to the best fitting SED in the rest-optical, we compile a sample of 241 BH*-dominated candidates from $\sim1000\,{\rm arcmin}^2$ of legacy and pure parallel JWST imaging. Our selection does not require a blue UV-component, and it successfully identifies objects that resemble the paradigmatic sources ``MoM-BH*-1'' and ``The Cliff''. We find that BH*-dominated sources exist across a wide range of redshifts ($z\sim1.7-9.3$) and optical luminosities (log$(L_{5100}/{\rm erg}\,{\rm s}^{-1})\sim42-44.5$), and we measure a median Balmer break strength of $\sim3$, with some breaks reaching values $>10$. We estimate bolometric luminosities in the range log$(L_{\rm bol}/{\rm erg}\,{\rm s}^{-1})\sim42-45$, which, assuming accretion at the Eddington-limit, would translate to black hole masses of $M_{\rm BH}\sim10^4-10^7{\rm M_\odot}$, spanning the intermediate mass black hole to the quasar regime. The number density of BH*-dominated candidates peaks at $z\sim5-6$ ($\sim10^{-5}\,{\rm Mpc}^{-3}$) and it declines by an order of magnitude down to $z\sim2$. Tentatively, comparing to V-shaped LRD samples suggests that the fraction of BH*-dominated sources among the broader LRD population does not decrease towards lower redshift. Crucially, our work demonstrates that BH*-dominated sources are not merely an early-Universe phenomenon but rather persist at least until cosmic noon.

astro-ph.GA↗

arXiv2026

Pristine Massive Star Formation Caught at the Break of Cosmic Dawn

The existence of galaxies with no heavy elements is a key prediction of cosmological models. So far no "zero-metallicity", or Population~III, galaxies have been identified. Here, we report the identification of an extremely metal-poor galaxy AMORE6 at redshift $z=5.7253$, multiply imaged by a foreground galaxy cluster. JWST spectra consistently detect ${\rm Hβ}$ at both positions, but [OIII]$_{λ\lambda4960,5008}$ remains undetected. This places a firm upper limit on its oxygen abundance, $<$0.19% of solar metallicity at 3$σ$, establishing itself as the most pristine galaxy by far. AMORE6 exhibits exceptional properties that indicate the presence of pristine massive star formation. Finding such an example at a relatively late cosmic time is surprising, but it also validates the basic ideas behind the Big Bang model.

astro-ph.CO↗

arXiv2026

Little Red Dot $-$ Host Galaxy $=$ Black Hole Star: A Gas-Enshrouded Heart at the Center of Every Little Red Dot

The central engines of Little Red Dots (LRDs) may be ``black hole stars" (BH*s), early stages of black hole growth characterized by dense gas envelopes. So far, the most direct evidence for BH*s comes from a handful of sources where the host galaxy is completely outshone as suggested by their remarkably steep Balmer breaks. Here we present a novel scheme to disentangle BH*s from their host galaxies assuming that the [OIII]5008Å line arises exclusively from the host. Using a sample of 98 LRDs ($z$~$2-9$) with high quality NIRSpec/PRISM spectra, we demonstrate that the host-subtracted median stack displays a Balmer break $>2\times$ stronger than massive quiescent galaxies, with the rest-optical continuum resembling a blackbody-like SED ($T_{\rm{eff}}$~$4050$ K, $\log(L_{\rm{bol}})$~$43.9$ erg s$^{-1}$, $R_{\rm{eff}}$~$1300$ au). We measure a steep Balmer decrement (H$α$/H$β>10$) and numerous density-sensitive features (e.g., FeII, HeI, OI). These are hallmark signatures of dense gas envelopes, providing population-level evidence that BH*s indeed power LRDs. In the median LRD, BH*s account for $\sim20\%$ of the UV emission, $\sim50\%$ at the Balmer break, and $\sim90\%$ at wavelengths longer than H$α$ with the remainder arising from the host. BH*s preferentially reside in low-mass galaxies ($M_{\rm{\star}}$~$10^{8}\,{\rm M}_{\rm{\odot}}$) undergoing recent starbursts, as evidenced by extreme emission line EWs (e.g., [OIII]5008Å~$1100$Å, CIII]~$12$Å), thereby favoring BH* origins linked to star-formation. We show V-shaped LRD selections are biased to high BH*/host fractions ($\gtrsim60\%$ at 5500Å) -- less dominant BH*s may be powering JWST's blue broad-line AGN. We find BH*s are so commonplace and transient (duty cycle $\sim1\%$, lifetime $\sim10$ Myrs) that every massive black hole may have once shone as a BH*.

astro-ph.GA↗

arXiv2026

How I Wonder What You Are -- JWST's Little Red Dots do not TWINKLE

Little Red Dots (LRDs) are a population of compact, red sources that have emerged as one of the most puzzling findings of JWST. Variability provides a direct probe of their central engines. Here we present the first joint spectroscopic and photometric time-domain study of LRDs undertaken with the JWST TWINKLE slitless spectroscopy program. Surveying the FRESCO GOODS-North legacy field, TWINKLE monitors a complete, H$α$-flux-limited sample of 18 LRDs at z = 3.9-6.8, achieving a rest-frame baseline of $\sim$140-220 days. We detect no variability in photometry, H$α$ line flux, or line shape across the sample. If LRDs resembled AGN in reverberation mapping samples -- the foundation for black hole mass calibrations and luminosity scaling relations -- we would expect >10 sources to show measurable fluctuations. Observing none implies a 5.9$σ$ deficit. The non-detections hold across all broad H$α$ emitters within TWINKLE's field of view -- the 18 V-shaped LRDs as well as 9 non-LRDs. Comparison with simulated light curves disfavors sub-Eddington accretion and is instead consistent with super-Eddington accretion, other mechanisms that suppress variability, or perhaps no AGN whatsoever. If LRDs do harbor black holes, calibrations derived from sub-Eddington systems may not apply, thereby explaining JWST's apparently "overmassive" black holes. These observations provide unique constraints on the physics of one of the most enigmatic populations discovered by JWST.

astro-ph.GA↗

arXiv2026

Entire Space Counterfactual Learning for Reliable Content Recommendations

Post-click conversion rate (CVR) estimation is a fundamental task in developing effective recommender systems, yet it faces challenges from data sparsity and sample selection bias. To handle both challenges, the entire space multitask models are employed to decompose the user behavior track into a sequence of exposure $\rightarrow$ click $\rightarrow$ conversion, constructing surrogate learning tasks for CVR estimation. However, these methods suffer from two significant defects: (1) intrinsic estimation bias (IEB), where the CVR estimates are higher than the actual values; (2) false independence prior (FIP), where the causal relationship between clicks and subsequent conversions is potentially overlooked. To overcome these limitations, we develop a model-agnostic framework, namely Entire Space Counterfactual Multitask Model (ESCM$^2$), which incorporates a counterfactual risk minimizer within the ESMM framework to regularize CVR estimation. Experiments conducted on large-scale industrial recommendation datasets and an online industrial recommendation service demonstrate that ESCM$^2$ effectively mitigates IEB and FIP defects and substantially enhances recommendation performance.

cs.LG↗

arXiv2026

Proximity Matters: Local Proximity Enhanced Balancing for Treatment Effect Estimation

Heterogeneous treatment effect (HTE) estimation from observational data poses significant challenges due to treatment selection bias. Existing methods address this bias by minimizing distribution discrepancies between treatment groups in latent space, focusing on global alignment. However, the fruitful aspect of local proximity, where similar units exhibit similar outcomes, is often overlooked. In this study, we propose Proximity-enhanced CounterFactual Regression (CFR-Pro) to exploit proximity for enhancing representation balancing within the HTE estimation context. Specifically, we introduce a pair-wise proximity regularizer based on optimal transport to incorporate the local proximity in discrepancy calculation. However, the curse of dimensionality renders the proximity measure and discrepancy estimation ineffective -- exacerbated by limited data availability for HTE estimation. To handle this problem, we further develop an informative subspace projector, which trades off minimal distance precision for improved sample complexity. Extensive experiments demonstrate that CFR-Pro accurately matches units across different treatment groups, effectively mitigates treatment selection bias, and significantly outperforms competitors. Code is available at https://github.com/HowardZJU/CFR-Pro.

cs.LG↗

arXiv2026

An Accurate and Interpretable Framework for Trustworthy Process Monitoring

Trustworthy process monitoring seeks to build an accurate and interpretable monitoring framework, which is critical for ensuring the safety of energy conversion plant (ECP) that operates under extreme working conditions such as high pressure and temperature. Contemporary self-attentive models, however, fall short in this domain for two main reasons. First, they rely on step-wise correlations that fail to involve physically meaningful semantics in ECP logs, resulting in suboptimal accuracy and interpretability. Second, attention matrices are frequently cluttered with spurious correlations that obscure physically meaningful ones, further impeding effective interpretation. To overcome these issues, we propose AttentionMixer, a framework aimed at improving both accuracy and interpretability of existing methods and establish a trustworthy ECP monitoring framework. Specifically, to tackle the first issue, we employ a spatial adaptive message passing block to capture variate-wise correlations. This block is coupled with a temporal adaptive message passing block through an \textit{mixing} operator, yielding a multi-faceted representation of ECP logs accounting for both step-wise and variate-wise correlations. Concurrently, to tackle the second issue, we employ a sparse message passing regularizer to filter out spurious correlations. We validate the efficacy of AttentionMixer using two real-world datasets from the radiation monitoring network for Chinese nuclear power plants.

cs.AI↗

arXiv2026

BEACON: JWST NIRCam Pure-parallel Imaging Survey. II. Physical Properties of $z=7-14$ Galaxies

We present photometric properties of 161 galaxy candidates at $z=7-14$ selected from the second data release (DR2) of BEACON, a JWST Cycle 2 pure-parallel NIRCam imaging program. Carefully selected from 36 independent pointings (corresponding to $\sim350$\,arcmin$^2$ sky coverage), and hence with reduced cosmic variance, our galaxy candidates provide an unbiased sample for investigating galaxy properties over a wide range of environments. We measure the physical properties, including UV continuum slope ($β_{\rm UV}$), stellar mass ($M_*$), star formation rate (SFR), and sizes. Our highest redshift galaxy candidate at $z=13.71\pm0.15$ has a remarkably bright UV luminosity of $M_{\rm UV}=-21.19\pm0.08$, making it the brightest galaxy at $z>12$ if spectroscopically confirmed. With an extremely blue UV slope, compact morphology, and high star formation rate surface density ($Σ_{\rm SFR}$), this candidate may have extremely low metallicity, high ionizing photon escape fraction, or contributions from an AGN. Among our multiple independent sightlines, we identify three fields of galaxy number overdensity with $>3σ$ significance. The properties of galaxies in various environments do not exhibit significant differences, implying either that accelerated galaxy evolution in overdense regions is not yet widespread at $z>7$, or that the current constraints are limited by sample size. Our simulations indicate that increasing the sample by an order of magnitude would allow such environmental trends to be robustly confirmed or ruled out, underscoring the importance of future pure-parallel observations.

astro-ph.GA↗

arXiv2026

Characterizing Dust Extinction and Spatially Resolved Paschen-$α$ Emission within 97 Galaxies at $1<z<1.6$ with JWST NIRCam Slitless Spectroscopy

We present results on the Paschen-$α$ (Pa$α$) emitting galaxies observed as part of the JWST FRESCO survey in the GOODS-North and GOODS-South fields. Utilizing the JWST NIRCam wide field slitless spectroscopy (WFSS), we analyze emission line fluxes, star formation rates (SFRs), and spatially resolved flux distributions of 97 Pa$α$ emitters at $1<z<1.6$. To assess dust extinction within our sample, we combine Pa$α$ fluxes with archival H$α$ data taken with the Hubble Space Telescope (HST) WFC3 G141 grism. Our analysis reveals a significant correlation between dust extinction and galaxy stellar mass, where more massive galaxies exhibit greater dust extinction. We employ two-dimensional Pa$α$ and F444W mapping to trace the distributions of star formation and stellar mass, respectively. Our observations indicate that lower mass galaxies are almost dust free in Pa$α$ and exhibit smaller sizes both in star formation and underlying stellar continuum. In contrast, galaxies with a stellar mass greater than $10^{9.5}M_\odot$ display diverse dust extinction and star formation patterns. This variation suggests that the structures and properties of massive galaxies evolve through different phases, which involve, e.g., star formation in massive clumps, compaction, and inside-out quenching. This study demonstrates the capabilities of JWST WFSS in conducting systematic investigations of emission line galaxies and highlights the pivotal role of Pa$α$ in advancing our understanding of dust extinction and obscured star formation in the early universe.

astro-ph.GA↗

arXiv2025

Environmental Regulation of Dust and Star Formation Unveiled by Subaru Dual Narrow-band Imaging: Degree-scale Balmer Decrement Mapping across a z = 0.9 Supercluster

We present results from a dual narrow-band imaging survey targeting the CL1604 supercluster at z = 0.9 using the Subaru Telescope. By combining the NB921 filter on HSC and the NB1244 filter on SWIMS, we can detect redshifted H$α$ and H$β$ emission lines from the supercluster. This unique technique allows us to measure both star formation rates and dust extinction for a sample of 94 emission-line galaxies across the supercluster. We find that dust extinction, estimated from the Balmer decrement (H$α$/H$β$ ratio), increases with stellar mass in star-forming galaxies, whereas relatively quiescent systems exhibit comparatively low extinction. Among galaxies with intermediate masses ($10^{8.5} < M_* < 10^{10.5}\,M_\odot$), the dust-corrected H$α$-based star formation rates align with the main sequence at this epoch. More massive galaxies, however, deviate from this relation, exhibit redder colors, and reside predominantly in higher-density environments. Although stellar mass, SFR, and galaxy color are clearly influenced by environment, we detect no strong, systematic environmental dependence of dust extinction for the whole sample.

astro-ph.GA↗

arXiv2025

Mixture of Low Rank Adaptation with Partial Parameter Sharing for Time Series Forecasting

Multi-task forecasting has become the standard approach for time-series forecasting (TSF). However, we show that it suffers from an Expressiveness Bottleneck, where predictions at different time steps share the same representation, leading to unavoidable errors even with optimal representations. To address this issue, we propose a two-stage framework: first, pre-train a foundation model for one-step-ahead prediction; then, adapt it using step-specific LoRA modules.This design enables the foundation model to handle any number of forecast steps while avoiding the expressiveness bottleneck. We further introduce the Mixture-of-LoRA (MoLA) model, which employs adaptively weighted LoRA experts to achieve partial parameter sharing across steps. This approach enhances both efficiency and forecasting performance by exploiting interdependencies between forecast steps. Experiments show that MoLA significantly improves model expressiveness and outperforms state-of-the-art time-series forecasting methods. Code is available at https://anonymous.4open.science/r/MoLA-BC92.

cs.LG↗

arXiv2025

Green Flash: Residual Emissions Enshrouded in Low-mass Balmer-break Galaxies at $z\sim5$

Recent James-Webb Space Telescope (JWST) observations have discovered galaxies that are already passively evolving at $z>4$, $\sim1.5$\,Gyr after the Big Bang. Remarkably, some of these galaxies exhibit strong emission lines such as \ha\ and \oiii\ while showing a strong continuum break at $\sim3650$\,Å i.e., Balmer break, giving us a unique insight into the physical mechanisms responsible for early galaxy quenching. In this study, we investigate the nature of four such galaxies at $z=5.10$--$5.78$ identified in the Abell~2744 field, using JWST/NIRCam and NIRSpec data. Our spectral energy distribution (SED) fitting analysis reveals that these galaxies have been mostly quiescent since $\sim100$\,Myr prior to the observed time. We find a higher dust attenuation in the nebular component than in the continuum in all cases. This suggests the presence of dusty star-forming regions or obscured AGN, which could be a {\it residual} signature of past quenching. For one of the galaxies with sufficient medium-band coverage, we derive the \hb+\oiii\ emission line map, finding that the line-emitting region is located in the center and is more compact ($R_e=0.7$\,kpc) than the stellar component ($R_e=0.9$\,kpc). For this specific galaxy, we discuss a scenario where quenching proceeds in the manner of ``outside-in", a stark contrast to the inside-out quenching commonly seen in massive galaxies at later cosmic times.

astro-ph.GA↗