cd /news/artificial-intelligence/holo-fuse-class-conditional-diffusio… · home topics artificial-intelligence article
[ARTICLE · art-56303] src=discuss.huggingface.co ↗ pub= topic=artificial-intelligence verified=true sentiment=· neutral

HoLo-FuSe — class-conditional diffusion on the 0-parameter HSL byte substrate (minimal-scale baseline, honest results)

HoLo-FuSe, a class-conditional diffusion model using a frozen 27-D HSL byte substrate with zero learned parameters, successfully generated cat and dog faces from AFHQ data at 128px resolution on a single Colab T4. The minimal-scale baseline run demonstrated proof of operation, matching the performance of a same-budget learned embedding control while using no learned conditioning parameters.

read2 min views1 publishedJul 12, 2026

Following up on the HoLo line (byte-LM, speech): HoLo-FuSe tests whether the same frozen HSL substrate can serve as the conditioning door of an image-generation carrier. Same method as the

other rooms: take a verified baseline, swap exactly one door for HSL, measure against controls.

Honest framing first. This was trained on a single free Colab T4 (~35M U-Net, 16k steps/arm,

128px). It is a minimum-scale baseline run — the point is proof of operation, not visual

quality. Please read the samples with that in mind; compute, not the method, is the main

quality ceiling here.

HSL is a frozen, deterministic 27-D feature frame over bytes (value geometry + cross-byte flow +

boundary + Fourier + phase; a 4.6 KB LUT, 0 learned parameters). Per the family rule —

fixed substrate where possible, explicit lens where necessary — the only lens this room needs is

a label→condition readout:

label bytes -> frozen 27-D HSL frame (0 params) -> small learned readout (2-layer MLP) -> condition embedding -> added to the DDPM timestep embedding No spatial lens: the conv U-Net carrier already owns spatial structure. The substrate stays frozen;

the readout is the only trained conditioning component, and it is budget-matched to the control.

Class-conditional DDPM (cosine schedule T=250, multi-level U-Net + self-attention, EMA, classifier-free guidance with cond-drop 0.15). Data: AFHQ animal faces at 128px, Cat 5153 / Dog

4739 (CC BY-NC 4.0 — so weights and samples are non-commercial). Three arms, same seed, same data, same architecture surface, same budget, compared at step 14000:

| arm | conditioning | result (qualitative, seed-matched) |
|---|---|---|

none | unconditional baseline | readable cat+dog faces, classes mixed | learned | same-budget nn.Embedding control | “Cat”→cats, “Dog”→dogs | hsl | frozen HSL 27-D (0 learned params) + readout | “Cat”→cats, “Dog”→dogs |

Two observations from the seed-matched bench (image attached; positions share the same initial

noise): The plan is to focus on quality of this one model (longer training, color balance, multi-seed

evaluation) rather than adding more models. That said, the compute budget is what it is — a free

T4 and a 4 GB laptop GPU — so please keep expectations modest. If the substrate claim survives

better-trained carriers, that is the result we are after; prettier cats are a bonus.

Code, training harness, bench scripts, and the POC record:

https://github.com/Woojiggun/HoLo-FuSe Live demo (ZeroGPU, generates in seconds; checkpoints are linked from the repo):

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @holo-fuse 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/holo-fuse-class-cond…] indexed:0 read:2min 2026-07-12 ·