Kinyugo Maina (@mkinyugo) header
Kinyugo Maina (@mkinyugo) profile picture

Machine Learning Researcher/Engineer. Working on cutting edge image and audio AI | ex @pixelcutai

Kenya
Joined May 2019
Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)Photo by Kinyugo Maina (@mkinyugo)
🔥High Fidelity Music Synthesis on a Shoestring Budget🔥 Hmm, that sounds like a pipe-dream🤔 Well it's not. I am excited to present my project, Msanii, a diffusion model for music synthesis that generates minutes of high fidelity audio efficiently.
9
20
3
106
21,843
There is no substitute for the person who Knows What To Do.
1,316
4,855
482
41,390
3,003,607
Kinyugo Maina retweeted
Replying to @renald_vie
there is no replacing a founder like Jensen. There never has been, and there can never be, a true succession for a founder like him. Nvidia ends today, and is replaced with a thing cosplaying nvidia. It will be run by custodians trying and failing to preserve the flame he ignited
299
1,148
194
14,885
1,379,353
Kinyugo Maina retweeted
This is a cool paper, but I think it's best thought of as Astra is surprisingly capable of doing old school mechanistic interpretability research rather than anything spooky. The things it's doing are clever, but typically exploiting structure in the data and using known method mechanistic interpretability techniques (which Astra was surely pretrained on), though it is better at this than I would have predicted! This is an exciting development for the prospects of ambitiously scaling interpretability or doing very laborious analysis like this, on more real models. However, I'm not holding my breath. This kind of analysis was pretty common up until about 2022, it largely fell out of favor because a lot of it is just exploiting fairly clean surface-level patterns that doesn't scale to the annoying complexity of real LLMs or which only works for the more simpler and superficial properties. And as far as I can tell, basically all the case studies here are just using a bag of interpretability tricks rather than anything spooky. The piece doesn't really update me on how far old school mech interp can go, just on how good LLMs are at it. It's very fun to see the creative solutions Astra came up with though!
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math. rikiyat.github.io/blog/weight-sp… github.com/RikiyaT/lm_und… huggingface.co/Qwen/Qwen3.5-0… huggingface.co/RikiyaT/Qwen3.…
52
164
37
1,866
106,625
Kinyugo Maina retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math. rikiyat.github.io/blog/weight-sp… github.com/RikiyaT/lm_und… huggingface.co/Qwen/Qwen3.5-0… huggingface.co/RikiyaT/Qwen3.…
2,368
7,254
5,755
55,917
10,091,258
Kinyugo Maina retweeted
The first inherently interpretable AI platform is finally here. Welcome to Clarity.
69
54
75
511
571,996
Kinyugo Maina retweeted
[1/2] Most image gen models output flat pixels. Users need layers. Our MRT (CVPR'2026) closes that gap and supports: • text → layers • image → layers • layers → layers 🔥 New SOTA on all three tasks 🔥 Beats Qwen-Image-Layered 🔥 10–100× faster, 50–90% less GPU memory
9
28
1
183
12,709
Kinyugo Maina retweeted
For over a decade, we’ve accepted that end-to-end backprop is the only way to train deep networks. But holding the entire network in memory all at once is why AI training is hitting a resource wall. We found a new way to break the network into blocks and train them independently. The trick? Treating the network’s forward pass like a diffusion model denoising a signal. This reinterpretation slashes the memory needed to train deep models. In our #ICLR2026 paper (arxiv.org/abs/2506.14202), we matched end-to-end performance across ViTs, DiTs, and LLMs. We did this while training just one isolated block at a time.
Introducing DiffusionBlocks: Block-wise Neural Network Training via Diffusion Interpretation pub.sakana.ai/diffusionblock… What if we didn’t have to hold an entire neural network in memory to train it? Standard neural net training optimizes all parameters jointly. As a result, the memory required during training grows linearly with the depth of the network. In our #ICLR2026 paper, we propose DiffusionBlocks, a principled framework to train networks one block at a time, drastically reducing memory requirements while matching end-to-end performance. With DiffusionBlocks, we split the network into blocks and train them one at a time, so you only need memory for a single block. How? We explicitly assign each block a role: to move the representation a little closer to the target than the block before it did. That role turns out to be precisely what a diffusion model does, step by step. Each block only needs to optimize its own objective and can be trained independently. We validated this across five different architectures: • ViT • DiT • Masked diffusion • Autoregressive transformers • Recurrent-depth transformers In each case, performance is competitive with end-to-end training while using a fraction of the memory. This perspective also extends naturally to recurrent-depth (Looped) transformers, which apply the same network iteratively and normally require expensive backpropagation through time (BPTT). Viewed through DiffusionBlocks, we can replace those multiple iterations with a single forward pass during training. Read our paper and code, to learn more. Paper: arxiv.org/abs/2506.14202 GitHub: github.com/SakanaAI/Diffusio… 🐟
150
614
79
5,689
759,280
Musings on Spairally Disclaimer - My background is in Computer Vision; therefore, I do not claim authoritative expertise in Audio Signal Processing. - My conclusions are drawn from a brief analysis of the product and the research presented by the founder.
1
1
1
147
4. Hardware Considerations The claim of upcoming hardware potentially improves the outlook for both concealed weapon detection and active shooter scenarios by bypassing the physical limitations of smartphones. This especially holds in an enterprise setting where optimized acoustic sensors can be installed. However, in the mobile setting, while custom wearables can incorporate more robust multi-mic spatial arrays, they still face challenges such as "acoustic shadowing" by the user's body and the limitations of TDOA accuracy caused by the compact form factor.
1
89
5. Conclusions Based on the available evidence, we cannot make definitive remarks on the feasibility or scalability of the product. The analysis above relies on technical heuristics; without a formal white paper, we can only form assumptions regarding the methodology. The global landscape of public safety is undergoing profound change, and it is impressive to see innovation emerging from the Kenyan tech ecosystem. If successful such a system is a meaningful contribution to making personal safety more accessible. I welcome corrections or alternative interpretations of the technical points I raised here, particularly from those with deep expertise in the relevant fields. My analysis is based on the publicly available information and a limited background in this specific domain. I look forward to the upcoming demo and hardware release, as well as any future publications or white paper that would enable further technical evaluation.
37
Kinyugo Maina retweeted
Diffusion models treat every part of an image equally. → Same number of steps. Same compute. But images aren’t uniform. 🤔 Some regions are easy, others are hard. So why force the model to treat them the same? 🧵
10
93
9
657
83,151
Kinyugo Maina retweeted
What if we could do Muon with just one Newton step while also achieving better loss?
6
22
3
222
29,165
Kinyugo Maina retweeted
Here's a short video from our founder, Zhilin Yang. (It's his first time speaking on camera like this, and he really wanted to share Kimi K2.5 with you!)
228
828
275
9,370
1,698,665
Kinyugo Maina retweeted
Today, we are open-sourcing Hunyuan World 1.1 (WorldMirror), a universal feed-forward 3D reconstruction model. 🚀🚀🚀   While our previously released Hunyuan World 1.0 (open-sourced, lite version deployable on consumer GPUs) focused on generating 3D worlds from text or single-view images, Hunyuan World 1.1 significantly expands the input scope by unlocking video-to-3D and multi-view-to-3D world creation.   Highlights: 🔹Any Input, Maximized Flexibility and Fidelity: Flexibly integrates diverse geometric priors (camera poses, intrinsics, depth maps) to resolve structural ambiguities and ensure geometrically consistent 3D outputs. 🔹Any Output, SOTA Results:This elegant architecture simultaneously generates multiple 3D representations: dense point clouds, multi-view depth maps, camera parameters, surface normals, and 3D Gaussian Splattings. 🔹Single-GPU & Fast Inference: As an all-in-one, feed-forward model, Hunyuan World 1.1 runs on a single GPU and delivers all 3D attributes in a single forward pass, within seconds.   🌐Project Page: 3d-models.hunyuan.tencent.co… 🔗Github:github.com/Tencent-Hunyuan/H… 🤗Hugging Face:huggingface.co/tencent/Hunyu… ✨Demo: huggingface.co/spaces/tencen… 📄Technical Report: 3d-models.hunyuan.tencent.co…
45
261
60
1,582
169,064
Kinyugo Maina retweeted
Media Invitation to our Press Conference
200
2,629
134
7,197
358,562
clutter on your desk, clutter on your mind
1
65
Kinyugo Maina retweeted
Google introduces Test-Time Diffusion Deep Researcher Don't sleep on diffusion models. Test-Time Diffusion Deep Researcher (TTD-DR) is a deep research agent that models research writing as a diffusion process. Instead of static reasoning or bolted-on tools, the system drafts an initial noisy report and iteratively refines it through retrieval and self-evolution, mimicking how humans plan, search, and revise. The approach depends on three stages: 1) generate a research plan 2) iteratively search with sub-agents that generate questions and synthesize answers (RAG-style) 3) compile findings into a final report. Multiple answer variants are created, scored by LLM judges, revised with feedback, and merged, yielding higher quality intermediate results. Draft reports are repeatedly revised using newly retrieved evidence, progressively improving accuracy and coherence until the final report is produced. On benchmarks like DeepConsult, Humanity’s Last Exam, and GAIA, TTD-DR beats OpenAI Deep Research by up to 74.5% win rates in long-form tasks and shows consistent gains in multi-hop reasoning. It's scalable, too! Ablation studies show each component adds measurable improvements, while Pareto analyses reveal better quality-latency tradeoffs than other DR agents. research.google/blog/deep-re…
19
178
6
1,045
104,198