iiinova — where your digital self keeps evolving

This project wouldn’t have been possible without this fantastic team! 🙌

000

The maximum SMS mass as the bright-end cutoff of the LRD luminosity function: Reconciling the low BH* masses with the bolometric luminosities from the fits requires Lbol/Ledd ~ 50. At this ratio, a maximal SMS’s luminosity matches the LRD luminosity function bright-end cutoff.

100

BH* masses are consistent with an SMS origin: All the derived BH* masses fall below the theoretical SMS mass ceiling, opening the possibility that a BH* arises from a single SMS. The low metallicities from the fits also place BH*s in the metal-poor environments where SMSs are expected to form.

100

BH*s are not “overmassive”: Measured against host stellar masses of 10⁸ M☉, our BH* masses are ~3 dex lower than estimates that report black holes orders of magnitude above the local scaling relation between black hole mass and stellar mass.

100

Based on these parameters, we explore four ways to estimating BH* masses, following Naidu et al. (2026). For the typical BH*, all of these methods yield consistent fiducial masses of 10⁴-10⁵ M☉. From the luminous to the faint BH* stacks, the masses span 10³·⁵-10⁶ M☉. This has several implications.

100

The model fitting yields a self-consistent set of parameters. In particular, the effective temperatures and bolometric luminosities place BH*s at the coldest yet most luminous corner of the HR diagram of known objects with photospheres or pseudo-photospheres.

100

We then fit tailored stellar atmosphere models from Liu et al. (2026) to BH*s.

100

To estimate BH* masses without contamination from their host galaxies, we first subtract the hosts to isolate the BH* spectra.

100

However, a basic challenge to the SMS scenario is that an SMS has a maximum mass of 10⁵-10⁶ M☉ imposed by general relativistic instability. Virtually all existing mass estimators, at face value, rule out this SMS origin.

100

Settling the mass debate is critical, as the BH* masses could discriminate between SMBH formation channels. Specifically, the spectral similarities between LRDs and massive star eruptions invite us to test one particular channel: Black Hole Stars may form from single supermassive stars.

100

Traditional calibrations, virial estimators based on broad Balmer lines, may not apply to LRDs, as LRDs differ from classical AGN in many ways.

100

The most basic of LRDs’ parameters – the masses of their central engines (“black hole stars,” BH*s) – remains debated by several orders of magnitude, from seed black holes to supermassive black holes.

100

International Space Science Institute (ISSI) @issibern.ch · 06/02/2026

Little Red Dots (LRDs) have unique properties, motivating new mass estimators for their central engines. We explore ways to derive their central engine masses. Our results are consistent with LRDs hosting black hole seeds as massive as single supermassive stars (SMSs). 🧵 arxiv.org/abs/2609.09274 🔭🧪

140

Reposted by Wendy Sun

This week, we are hosting ISSI International Team 659, “Little Red Dots, Big Open Questions”, led by Mengyuan Xiao & Rohan Naidu @rpnaidu.bsky.social. The team brings together observers and theorists to study one of the James Webb Space Telescope’s most intriguing discoveries: Little Red Dots. 🔴🔭

The images show examples of Little Red Dots observed by JWST, highlighting their compact structure and distinctive red colours in the early Universe.

Image Credit: © NASA/CSA/ESA, P. A. Oesch & M. Xiao (University of Geneva), J. Matthee (ISTA), G. Brammer (Niels Bohr Institute)

062

This project wouldn’t have been possible without this incredible team, featuring @rpnaidu.bsky.social @jorryt.bsky.social, and Anna de Graaff! 🙌

060

We also infer a BH* duty cycle of ~1% and a BH* lifetime of ~10 Myrs - they are a short-lived, and yet ubiquitous phenomenon. They may be a key phase in the origin story of possibly every massive BH!

130

V-shaped LRD selections are preferentially sensitive to high BH*/LRD fractions. So the population of BH*s may be more widespread than what we currently know. Hidden BH*s in e.g. broad-line AGN??

With BH*-dominated LRDs on one extreme ⬅️, the other extreme has host-dominated LRDs ➡️, which appear to be some of the most spectacular starbursts at high redshift.

BH*/LRD fractions correlate with easily accessible observables, LRD [OIII] EW and Balmer break strength. This may help accelerate the search for BH*-dominated objects.

Who dominates the LRD light? In the UV, the BH* contribution is modest relative to the host. But then there is an abrupt transition around (but not exactly at) the Balmer break, redwards of which the BH* begins to dominate the light.

LRD hosts lie above the star-forming main sequence, and display rising star-formation histories. Recent starbursts may play a role in the formation of BH*s and/or their entrapment in gas!

What about the LRD hosts? They show far stronger emission lines than a control sample, implying highly ionizing, young stellar populations.

All the features (in more detail in the paper) lead us to conclude that BH*s are indeed the central engines of LRDs!

120

When we look at prominent LRDs like The Cliff and MoM-BH*-1 (~“pure” BH*), they bear remarkable similarities to the (LRD - host) stack. And A2744-45924 shows several emission features (e.g., Fe II-UV) that are also apparent in the stack.

210

A single blackbody with Teff ~4000 K fits the rest-optical well. The inferred radius ~1300 au is 2 dex larger than the largest known stars, but comparable to local broad-line regions.

The stack has a strong Balmer break ~6.50 that far exceeds the strongest breaks seen in quiescent galaxies, as well as EW(Hα) ~850Å and Balmer decrement ~16 that are far more extreme than local AGN!

The (LRD - host) central engine stack exhibits a constellation of features that are the hallmark signatures of BH*s!

120

We present a novel, empirical approach to isolate the central engines of LRDs, by assuming that the [OIII] luminosity arises purely from the host galaxy. By subtracting the host, we uncover what’s remaining. LRD - Host Galaxy = ??

Wendy Sun @wendysun.bsky.social · 31/01/2026

Paper day!! Little Red Dots (LRDs) are seen everywhere by JWST, but even after 3 years of relentless effort we are still debating what these things actually are. We decompose LRDs to show LRD - Host Galaxy = Black Hole Star (BH*)! 🧵 arxiv.org/abs/2601.20929 🔭 🧪

3326

I'm an undergrad student at MIT, working in astronomy