arXiv subjects

Zihao Wu

Publications and source records attributed to Zihao Wu.

At least 55 records · Page 3Linked to original sources

arXiv2026

JWST Advanced Deep Extragalactic Survey (JADES) Data Release 5: Photometric Catalog

JADES Data Release 5 (DR5) photometric catalogs and describes the methodologies used for source detection, deblending, photometry, uncertainty estimation, and catalog curation. The catalogs are constructed from 35 space-based imaging mosaics obtained with JWST/NIRCam, JWST/MIRI, HST/ACS, and HST/WFC3, combining approximately 1250 hours of JADES imaging with extensive additional public JWST and HST observations in the GOODS fields. Sources are identified using custom signal-to-noise-based detection and deblending algorithms optimized for the depth, resolution, and complex point-spread-function structure of JWST imaging. Source centroids, shapes, and photometric apertures are determined using a new fast two-dimensional Gaussian regression method applied to detection-image profiles. We provide forced circular-aperture photometry, ellipsoidal Kron photometry, and curve-of-growth measurements for every source in every band. We introduce a new pixel-level regression framework to model photometric uncertainties as a function of aperture size and local mosaic properties, accounting for correlated noise in heterogeneous JWST mosaics. Photometric redshifts are computed using template-based fitting applied to both small-aperture photometry on unconvolved images and Kron photometry on common-PSF mosaics. The JADES DR5 catalogs supersede previous JADES photometric releases, and are publicly released through the Mikulski Archive for Space Telescopes and an interactive web interface.

astro-ph.GA↗

arXiv2026

JWST Advanced Deep Extragalactic Survey (JADES) Data Release 5: Catalogs of inferred morphological properties of galaxies from JWST/NIRCam imaging in GOODS-N and GOODS-S

We present morphological parameters and their uncertainties for all sources detected in JWST/NIRCam imaging in GOODS-N and GOODS-S from the JWST Advanced Deep Extragalactic Survey (JADES) catalogs. We model the surface brightness profiles of these sources with single-component Sérsic profiles, performing Bayesian inference of galaxy structural parameters. We fit each of the $>10^5$ sources with every available JWST/NIRCam wide-band filter individually, amounting to over 3 million Sérsic profiles computed. We provide catalogs of this morphological information, building one of the largest extragalactic morphological datasets to date, which we share alongside imaging and photometry from the JADES Data Release 5. With this information, we analyze the rest-frame optical redshift evolution of the effective radius and the surface luminosity density within a radius of 1 kiloparsec, $Σ_{\text{1 kpc}}$, for 24,692 galaxies at $z>1$. We find $r_{\text{eff}} \propto (1+z)^{-0.635 \pm 0.013}$ kpc, while $Σ_{\text{1 kpc}}$ is relatively constant across time. Additionally, we explore bulge-disk decomposition on a subset of 8,390 galaxies in the JADES deep imaging covering the Hubble Ultra Deep Field, finding the effective radius of the bulge-components to increase marginally with time, whereas the disk-component sizes evolve as $r_{\text{eff,disk}} \propto (1+z)^{-1.091 \pm 0.043}$. Future work modeling multi-component surface brightness profiles will enable further analysis of the morphological evolution of galaxies across cosmic time.

astro-ph.GA↗

arXiv2026

JWST Advanced Deep Extragalactic Survey (JADES) Data Release 5: Wisp Subtraction with the Non-negative Matrix Factorization Algorithm

Wisps are among the most prominent scattered light artifacts in JWST/NIRCam imaging. They often appear in certain regions of the detectors and contaminate observations at surface-brightness levels relevant for faint-source photometry. We introduce a new subtraction method that uses the non-negative matrix factorization (NMF) algorithm to model and remove wisps. Using deep NIRCam observations from the JWST Advanced Deep Extragalactic Survey (JADES) and other programs, we construct multi-component, filter- and detector-specific wisp templates that capture the wisp structures and their exposure-to-exposure morphological variations. Wisps in individual exposures are represented as non-negative linear combinations of these templates, consistent with their additive nature and reducing degeneracies relative to single-template scaling. Compared to existing approaches, our method delivers lower residual root mean square in wisp-affected regions and reduces photometric bias and scatter to levels consistent with clean detector areas. The NMF wisp templates are readily applicable to other datasets and are publicly released to support future NIRCam extragalactic surveys.

astro-ph.IM↗

arXiv2026

JADES: Discovery of Large Reservoirs of Small Dust Grains in the Circumgalactic Medium of Massive Galaxies at $z\sim3.5$ through Deep JWST/NIRCam Imaging and Grism Spectroscopy

Using JWST NIRCam imaging and grism spectroscopy from the JWST Advanced Deep Extragalactic Survey (JADES) Origins Fields, we report spectroscopic redshift measurements of 1,445 emission-line galaxies at $z=0-9$. Within this sample, we identify two prominent galaxy protoclusters at $z = 3.47$ and 3.69, each anchored by massive dusty star-forming galaxies (DSFGs). In the vicinity of these systems, we discover seven background galaxies at $z=3.6 - 6$ that simultaneously exhibit strong rest-frame optical emission lines (e.g., [O III] and H$α$) and unusually reddened UV-to-optical continua. We attribute this reddening to dust extinction arising from the circumgalactic medium (CGM) of the foreground DSFGs at projected separations of 7-30 kpc. We infer a high dust column density ($\gtrsim 10^{-1}$ Msun/kpc^2), substantially exceeding those measured in low-redshift halos and those predicted by hydrodynamical simulations like IllustrisTNG and FIRE-2. The steep extinction curves, comparable to or steeper than that of the SMC, indicate a dominant population of small dust grains in the high-redshift CGM. We conclude that DSFGs at this epoch host large reservoirs of dusty CGM enriched to solar metallicity. These extended dust components are largely invisible to (sub-)millimeter interferometers such as ALMA because of their low surface brightness. We discuss the physical processes in dust transport that might be key to reproducing our observations, including galaxy mergers, cool-phase gas outflows, dust shattering, sputtering and radiation pressure. Finally, we caution that foreground CGM dust extinction may redden background galaxies at intermediate redshifts to mimic Lyman-break galaxies at $z\gtrsim10$.

astro-ph.GA↗

arXiv2026

On the origins of oxygen: ALMA and JWST characterise the multi-phase, metal-enriched, star-bursting medium within a 'normal' $z > 11$ galaxy

The unexpectedly high abundance of galaxies at $z > 11$ revealed by JWST has sparked a debate on the nature of early galaxies and the physical mechanisms regulating their formation. The Atacama Large Millimeter/submillimeter Array (ALMA) has begun to provide vital insights on their gas and dust content, but so far only for extreme 'blue monsters'. Here we present new, deep ALMA observations of JADES-GS-z11-0, a more typical (sub-$L^*$) $z > 11$ galaxy that bridges the discovery space of JWST and the Hubble Space Telescope. These data confirm the presence of the [O III] 88 $μ$m line at $4.5σ$ significance, precisely at the redshift of several faint emission lines previously seen with JWST/NIRSpec, while the underlying dust continuum remains undetected ($F_ν< 9.0 \, \mathrm{μJy}$), implying an obscured star formation rate (SFR) of $\text{SFR}_\text{IR} \lesssim 6 \, \mathrm{M_\odot \, yr^{-1}}$ and dust mass of $M_\text{dust} \lesssim 1.0 \times 10^{6} \, \mathrm{M_\odot}$ (all $3σ$). The accurate ALMA redshift of $z_\text{[O III]} = 11.1221 \pm 0.0006$ ($\gtrsim \! 5\times$ refined over NIRSpec) helps confirm that redshifts measured purely from the Lyman-$α$ break, even spectroscopically, should properly take into account the effects of potential damped Lyman-$α$ absorption (DLA) systems to avoid systematic overestimates of up to $Δz \approx 0.5$. The [O III] 88 $μ$m luminosity of $L_\text{[O III]} = (1.1 \pm 0.3) \times 10^{8} \, \mathrm{L_\odot}$, meanwhile, agrees well with the scaling relation for local metal-poor dwarfs given the SFR measured by NIRCam, NIRSpec, and MIRI. The spatially resolved MIRI and ALMA emission also underscores that JADES-GS-z11-0 is likely to consist of two low-mass components that are undergoing strong bursts of star formation yet are already pre-enriched in oxygen (~20-30% solar), only 400 Myr after the Big Bang.

astro-ph.GA↗

arXiv2026

Brain-Inspired Exploration of Functional Networks and Key Neurons in Large Language Models

In recent years, the rapid advancement of large language models (LLMs) in natural language processing has sparked significant interest among researchers to understand their mechanisms and functional characteristics. Although prior studies have attempted to explain LLM functionalities by identifying and interpreting specific neurons, these efforts mostly focus on individual neuron contributions, neglecting the fact that human brain functions are realized through intricate interaction networks. Inspired by research on functional brain networks (FBNs) in the field of neuroscience, we utilize similar methodologies estabilished in FBN analysis to explore the "functional networks" within LLMs in this study. Experimental results highlight that, much like the human brain, LLMs exhibit certain functional networks that recur frequently during their operation. Further investigation reveals that these functional networks are indispensable for LLM performance. Inhibiting key functional networks severely impairs the model's capabilities. Conversely, amplifying the activity of neurons within these networks can enhance either the model's overall performance or its performance on specific tasks. This suggests that these functional networks are strongly associated with either specific tasks or the overall performance of the LLM. Code is available at https://github.com/WhatAboutMyStar/LLM_ACTIVATION.

q-bio.NC↗

arXiv2025

Ionizing Photon Production Efficiencies and Chemical Abundances at Cosmic Dawn Revealed by Ultra-Deep Rest-Frame Optical Spectroscopy of JADES-GS-z14-0

JWST has discovered an early period of galaxy formation that was more vigorous than expected, which has challenged our understanding of the early Universe. In this work, we present the longest spectroscopic integration ever acquired by JWST/MIRI. This spectrum covers the brightest rest-frame optical nebular emission lines for the luminous galaxy JADES-GS-z14-0 at $z > 14$. Most notably, we detect $[\mathrm{OIII}] λλ4959,5007$ at $\approx 11 σ$ and $\mathrm{H}α$ at $\approx 4 σ$ with these ultra-deep observations. These lines reveal that JADES-GS-z14-0 has low dust attenuation with a recent star-formation rate of $\mathrm{SFR} \approx 10 \pm 2\ M_{\odot} / \mathrm{yr}$, star-formation rate surface density of $Σ_{\mathrm{SFR}} \approx 23 \pm 5\ M_{\odot}/\mathrm{yr}/\mathrm{kpc}^{2}$, and ionizing photon production efficiency of $ξ_{\mathrm{ion}} \approx 10^{25.3 \pm 0.1}\ \mathrm{Hz/erg}$. Using standard strong-line diagnostics, we infer a gas-phase oxygen abundance of $[\mathrm{O/H}] \approx -1.1 \pm 0.4$ ($\approx 10\%\ Z_{\odot}$), carbon-to-oxygen ratio of $[\mathrm{C/O}] \approx -0.4 \pm 0.4$, ionization parameter of $\mathrm{log}_{10}(U) \gtrsim -2.4$, and density of $n_{\mathrm{H}} \approx 720 \pm 210\ \mathrm{cm}^{-3}$. Using detailed photoionization modeling, we instead derive $[\mathrm{O/H}] \approx -0.3_{-0.4}^{+0.4}$ ($\approx 50\%\ Z_{\odot}$) and $\mathrm{log}_{10}(U) \approx -1.5_{-0.4}^{+0.3}$. The inferred properties of JADES-GS-z14-0 are similar to those measured for similarly luminous galaxies at $z > 10$ with previous MIRI/Spectroscopy, such as GHZ2/GLASSz12, GN-z11, and MACS0647-JD1. Existing simulations are unable to reproduce the empirical and inferred properties of JADES-GS-z14-0. This work demonstrates an important step toward understanding the formation of the first stars and heavy elements in the Universe. [Abridged]

astro-ph.GA↗

arXiv2025

Complete computation of all three-loop five-point massless planar integrals

We calculate all three-loop, five-point, massless planar Feynman integral families in the dimensional regularization scheme. This is a new milestone in Feynman integral computations. The analysis covers four distinct families of Feynman integrals for this configuration, for all of which we derive the canonical differential equations. Our results also confirm a prediction on the three-loop five-point alphabet. The boundary values are analytically determined. Using these differential equations, the integrals can be evaluated to high precision efficiently. Our work establishes the foundation for next-to-next-to-next-to-leading-order (N$^3$LO) calculation of the production of three massless final states, as well as corresponding bootstrap studies in gauge theories.

hep-ph↗

arXiv2025

Little red dot variability over a century reveals black hole envelope via a giant Einstein cross

"Little red dots" (LRDs) represent a new population of astronomical objects uncovered by JWST whose nature remains debated. Although many LRDs are suspected as active galactic nuclei (AGN), they show little variability on days-years timescales. We report the discovery of two gravitationally lensed LRDs at redshift $\sim$4.3 behind the cluster RXCJ2211-0350, one of which (RX1) is quadruply imaged with time delays spanning $\sim$130 years. RX1 exhibits intrinsic color and brightness variations of up to 0.7 magnitude among its images. These changes are consistent with blackbody-temperature variations of a photosphere, indicating long-term variability analogous to Cepheid-like pulsations but in a far more extended ($R \sim 2000$ AU) and massive ($M \gtrsim 10^6 \, M_{\odot}$) systems. These results suggest LRDs as a distinct class of AGN with stellar-like envelopes.

astro-ph.GA↗

arXiv2025

DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4

The digitization of healthcare has facilitated the sharing and re-using of medical data but has also raised concerns about confidentiality and privacy. HIPAA (Health Insurance Portability and Accountability Act) mandates removing re-identifying information before the dissemination of medical records. Thus, effective and efficient solutions for de-identifying medical data, especially those in free-text forms, are highly needed. While various computer-assisted de-identification methods, including both rule-based and learning-based, have been developed and used in prior practice, such solutions still lack generalizability or need to be fine-tuned according to different scenarios, significantly imposing restrictions in wider use. The advancement of large language models (LLM), such as ChatGPT and GPT-4, have shown great potential in processing text data in the medical domain with zero-shot in-context learning, especially in the task of privacy protection, as these models can identify confidential information by their powerful named entity recognition (NER) capability. In this work, we developed a novel GPT4-enabled de-identification framework (``DeID-GPT") to automatically identify and remove the identifying information. Compared to existing commonly used medical text data de-identification methods, our developed DeID-GPT showed the highest accuracy and remarkable reliability in masking private information from the unstructured medical text while preserving the original structure and meaning of the text. This study is one of the earliest to utilize ChatGPT and GPT-4 for medical text data processing and de-identification, which provides insights for further research and solution development on the use of LLMs such as ChatGPT/GPT-4 in healthcare. Codes and benchmarking data information are available at https://github.com/yhydhx/ChatGPT-API.

cs.CL↗

arXiv2025

Evaluating Large Language Models for Radiology Natural Language Processing

The rise of large language models (LLMs) has marked a pivotal shift in the field of natural language processing (NLP). LLMs have revolutionized a multitude of domains, and they have made a significant impact in the medical field. Large language models are now more abundant than ever, and many of these models exhibit bilingual capabilities, proficient in both English and Chinese. However, a comprehensive evaluation of these models remains to be conducted. This lack of assessment is especially apparent within the context of radiology NLP. This study seeks to bridge this gap by critically evaluating thirty two LLMs in interpreting radiology reports, a crucial component of radiology NLP. Specifically, the ability to derive impressions from radiologic findings is assessed. The outcomes of this evaluation provide key insights into the performance, strengths, and weaknesses of these LLMs, informing their practical applications within the medical domain.

cs.CL↗

arXiv2025

GP-GPT: Large Language Model for Gene-Phenotype Mapping

Pre-trained large language models(LLMs) have attracted increasing attention in biomedical domains due to their success in natural language processing. However, the complex traits and heterogeneity of multi-sources genomics data pose significant challenges when adapting these models to the bioinformatics and biomedical field. To address these challenges, we present GP-GPT, the first specialized large language model for genetic-phenotype knowledge representation and genomics relation analysis. Our model is fine-tuned in two stages on a comprehensive corpus composed of over 3,000,000 terms in genomics, proteomics, and medical genetics, derived from multiple large-scale validated datasets and scientific publications. GP-GPT demonstrates proficiency in accurately retrieving medical genetics information and performing common genomics analysis tasks, such as genomics information retrieval and relationship determination. Comparative experiments across domain-specific tasks reveal that GP-GPT outperforms state-of-the-art LLMs, including Llama2, Llama3 and GPT-4. These results highlight GP-GPT's potential to enhance genetic disease relation research and facilitate accurate and efficient analysis in the fields of genomics and medical genetics. Our investigation demonstrated the subtle changes of bio-factor entities' representations in the GP-GPT, which suggested the opportunities for the application of LLMs to advancing gene-phenotype research.

cs.CL↗

arXiv2025

MoMoE: A Mixture of Expert Agent Model for Financial Sentiment Analysis

We present a novel approach called Mixture of Mixture of Expert (MoMoE) that combines the strengths of Mixture-of-Experts (MoE) architectures with collaborative multi-agent frameworks. By modifying the LLaMA 3.1 8B architecture to incorporate MoE layers in each agent of a layered collaborative structure, we create an ensemble of specialized expert agents that iteratively refine their outputs. Each agent leverages an MoE layer in its final attention block, enabling efficient task decomposition while maintaining computational feasibility. This hybrid approach creates specialized pathways through both the model architecture and the agent collaboration layers. Experimental results demonstrate significant improvements across multiple language understanding and generation benchmarks, highlighting the synergistic benefits of combining expert routing at both the neural and agent levels.

cs.CE↗

arXiv2025

Input Domain Aware MoE: Decoupling Routing Decisions from Task Optimization in Mixture of Experts

Sparse Mixture of Experts (sMoE) has become a pivotal approach for scaling large vision-language models, offering substantial capacity while maintaining computational efficiency through dynamic, sparse activation of experts. However, existing routing mechanisms, typically based on similarity scoring, struggle to effectively capture the underlying input structure. This limitation leads to a trade-off between expert specialization and balanced computation, hindering both scalability and performance. We propose Input Domain Aware MoE, a novel routing framework that leverages a probabilistic mixture model to better partition the input space. By modeling routing probabilities as a mixture of distributions, our method enables experts to develop clear specialization boundaries while achieving balanced utilization. Unlike conventional approaches, our routing mechanism is trained independently of task-specific objectives, allowing for stable optimization and decisive expert assignments. Empirical results on vision-language tasks demonstrate that our method consistently outperforms existing sMoE approaches, achieving higher task performance and improved expert utilization balance.

cs.LG↗

arXiv2025

The dark side of early galaxies: $\texttt{geko}$ uncovers dark-matter fractions at $z\sim4-6$

JWST/NIRCam slitless spectroscopy enables dynamical mass measurements for typical star-forming galaxies only a billion years after the Big Bang. We model the H$α$ morpho-kinematics of 163 galaxies at redshift $z\approx4$-6 from FRESCO and CONGRESS (with JADES imaging), using the $\texttt{geko}$ code, and infer rotational velocities and dispersions within $r_{\rm e}$. Our sample spans $\log M_{\star}\approx7$-10 and $\log M_{\rm dyn}\approx9$-11. Gas masses are estimated via scaling relations, yielding baryonic masses and dark-matter (DM) fractions $f_{\rm DM}(r<r_{\rm e})$ within the H$α$ half-light radius. We find high median fractions of $\langle f_{\rm gas}\rangle=0.77$ and $\langle f_{\rm DM}\rangle=0.73$, where $f_{\rm gas}$ is measured with respect to the baryonic mass and $f_{\rm DM}$ with respect to the DM+baryonic mass. About two-thirds of systems are DM-dominated within $r_{\rm e}\sim0.5-1$ kpc. Both $f_{\rm gas}$ and $f_{\rm DM}$ decrease with stellar mass, consistent with simulations. The stellar Tully-Fisher relation shows a tentative offset to higher $v_{\rm circ}$ at fixed $M_{\star}$ and substantial intrinsic scatter, suggesting that the relation is only beginning to emerge at $z\sim5$. We measure a negative correlation between $f_{\rm DM}$ and baryonic surface density $Σ_{\rm bar}$, weaker but broadly consistent with trends at cosmic noon and at $z\sim0$. Qualitatively comparing with modified NFW profiles coupled to an empirical stellar-to-halo mass relation suggests that the lowest $f_{\rm DM}$ ($\lesssim0.4$) require cored inner DM profiles, while the highest fractions favour cuspier profiles, potentially reflecting adiabatic contraction. Overall, the elevated $f_{\rm gas}$ and $f_{\rm DM}$ at $z\gtrsim4$ are compatible with progenitors of baryon-dominated systems at $z\sim2$ and naturally anticipate overmassive black holes at fixed $M_{\star}$.

astro-ph.GA↗

arXiv2025

Cloudy with a chance of starshine: Possible photometric signatures of nebular-dominated emission in $1.5 < z < 8.5$ JADES galaxies

The discovery of high-redshift galaxies exhibiting a steep spectral UV downturn potentially indicative of two-photon continuum emission marks a turning point in our search for signatures of top-heavy star formation in the early Universe. We develop a photometric search method for identifying further nebular-dominated galaxy candidates, whose nebular continuum dominates over the starlight, due to the high ionising photon production efficiencies $ξ_\mathrm{ion}$ associated with massive star formation. We utilise the extensive medium-band imaging from JADES, which enables the identification of Balmer jumps across a wide range of redshifts ($1.5 < z < 8.5$), through the deficit in rest-frame optical continuum level. As Balmer jumps are a general recombination feature of young starbursts ($\lesssim 3$~Myr), we further demand a high observed $\log\, (ξ_\mathrm{ion, obs}/\mathrm{(Hz\ erg^{-1})}) > 25.60$ to power the strong nebular continuum, together with a relatively non-blue UV slope indicating a lack of stellar continuum emission. Our nebular-dominated candidates, constituting ${\sim}$10% of galaxies at $z \sim 6$ (decreasing to ${\sim}$3% at $z \sim 2$, not completeness-corrected) are faint in the rest-frame optical (median $M_\mathrm{opt} = -17.95$) with extreme line emission (median $\mathrm{EW}_\mathrm{Hα,rest} = 1567$ Å, $\mathrm{EW}_\mathrm{[O\ III] + Hβ,rest} = 2244$ Å). However, hot H II region temperatures, collisionally-enhanced two-photon continuum emission, and strong UV lines are expected to accompany top-heavy star formation. Thus nebular-dominated galaxies do not necessarily exhibit the biggest Balmer jumps, nor the largest $ξ_\mathrm{ion, obs}$ or reddest UV slopes. Hence continuum spectroscopy is ultimately required to establish the presence of a two-photon downturn in our candidates, thus advancing our understanding of primordial star formation and AGN.

astro-ph.GA↗

arXiv2025

Instance-Aware Robust Consistency Regularization for Semi-Supervised Nuclei Instance Segmentation

Nuclei instance segmentation in pathological images is crucial for downstream tasks such as tumor microenvironment analysis. However, the high cost and scarcity of annotated data limit the applicability of fully supervised methods, while existing semi-supervised methods fail to adequately regularize consistency at the instance level, lack leverage of the inherent prior knowledge of pathological structures, and are prone to introducing noisy pseudo-labels during training. In this paper, we propose an Instance-Aware Robust Consistency Regularization Network (IRCR-Net) for accurate instance-level nuclei segmentation. Specifically, we introduce the Matching-Driven Instance-Aware Consistency (MIAC) and Prior-Driven Instance-Aware Consistency (PIAC) mechanisms to refine the nuclei instance segmentation result of the teacher and student subnetwork, particularly for densely distributed and overlapping nuclei. We incorporate morphological prior knowledge of nuclei in pathological images and utilize these priors to assess the quality of pseudo-labels generated from unlabeled data. Low-quality pseudo-labels are discarded, while high-quality predictions are enhanced to reduce pseudo-label noise and benefit the network's robust training. Experimental results demonstrate that the proposed method significantly enhances semi-supervised nuclei instance segmentation performance across multiple public datasets compared to existing approaches, even surpassing fully supervised methods in some scenarios.

cs.CV↗

arXiv2025

Beyond the stars: Linking H$α$ sizes, kinematics, and star formation in galaxies at $z\approx 4-6$ with JWST grism surveys and $\texttt{geko}$

Understanding how galaxies assemble their mass during the first billion years of cosmic time is a central goal of extragalactic astrophysics, yet joint constraints on their sizes and kinematics remain scarce. We present one of the first statistical studies of the $\mathrm{H}α$ size-mass relation at high redshift with a sample of 213 galaxies at spectroscopic redshifts of $z\approx 4-6$ from the FRESCO and CONGRESS NIRCam grism surveys. We measure the $\mathrm{H}α$ morphology and kinematics of our sample using the novel forward modelling Bayesian inference tool $\texttt{geko}$, and complement them with stellar continuum sizes in the rest-frame FUV, NUV, and optical, obtained from modelling of imaging data from the JADES survey with $\texttt{Pysersic}$. At $z\approx5$, we find that the average H$α$ sizes are larger than the stellar continuum (FUV, NUV and optical), with $r_{\rm e, Hα}= 1.17 \pm 0.05$ kpc and $r_{\rm e,cont} \approx 0.9$ kpc for galaxies with $\log(M_{\star} ~\rm [M_{\odot}])= 9.5$. However, we find no significant differences between the stellar continuum sizes at different wavelengths, suggesting that galaxies are not yet steadily growing inside-out at these epochs. Instead, we find that the ratio $r_{\rm e, Hα}/r_{\rm e, NUV}$ increases with the distance above the star-forming main sequence ($Δ\rm MS$), consistent with an expansion of H$α$ sizes during episodes of enhanced star formation caused by an increase in ionising photons. As galaxies move above the star-forming main sequence, we find an increase of their rotational support $v/σ$, which could be tracing accreting gas illuminated by the \Ha\ emission. Finally, we find that about half of the elongated systems ($b/a < 0.5$) are not rotationally supported, indicating a potential flattened/prolate galaxy population at high redshift.

astro-ph.GA↗