#SpatialAI

The Timeless Traveler 🇺🇸 @TimelessTrvlr

The Dolomites in northern Italy are unlike anywhere else in Europe. Towering limestone peaks rise above small alpine villages, green valleys, and winding mountain roads that make the drive just as memorable as the destination.

NVIDIA AI @NVIDIAAI

Code is the right action interface for spatial reasoning agents. New from NVIDIA Research: SpatialClaw, a training-free agent that uses code as its action interface for complex visual tasks. Instead of calling a fixed set of pre-defined tools, the agent writes Python inside a persistent kernel, so it can compose perception modules, inspect intermediate results, and revise its strategy across steps. Perception outputs become ordinary variables it can reuse and combine with libraries like NumPy and SciPy. With no benchmark-specific or model-specific tuning, it beats a recent prior agent by 11.2 points across 20 benchmarks and holds up consistently across six different model backbones. You can check out SpatialClaw here:

tagSpace | Find what people left at real places @tagSpaceCo

Layers are live. Switch on what you care about, switch off the rest.

Machine Delusions @Machinedelusion

L3G-World (KAIST): a world model whose memory is 3D Gaussians holding VAE latents, so it renders straight into the video model's latent space. Memory grows with the scene, not clip length. Real-time, 9 fps. Code coming.

Francois Chaubard @FrancoisChauba1

I believe that LSTM based architectures (constant hidden state, constant flops / step) will crush transformers as we know them in 2-3 years. we will laugh about KV Caches and how dumb we were.. to achieve this, the LSTM will need to have a large external memory bank with tool-use like Mem0 / RLMs / RAG to read/write memories into a db, but this will be more like us writing things in a notebook vs. being forced to memorize everything in a KV-cache. this will require the model NOT to memorize facts inside the weights, but ONLY learn functions to manipulate, write, and retrieve facts, which of course is the better system. if I am vibe coding, why do I need to know who Kevin Costner is, and how many people are in paris, and have all of those facts in my weight file.. seems like a waste. and the whole model + fixed hidden state can sit in SRAM and the "memories" will sit in cpu ram (or disk?), as it will be addressed in small amounts and perhaps even regex-able vs. big heavy knn mat mul. what is distinct about this structure is the model itself will need to learn a saliency function of what to store in short term, long term, vs. external memory such that it can be easily retrieved, and what to forget entirely... without ANY human example traces. just via RL. the biggest implication as it comes to chips and DRAM is that it will prefer chips w large SRAM and a huge amount of SMs and likely little DRAM if any at all. so chips that support this will be good investments. does anyone else see this yet?

𝕏 Travels, Hotels and Resorts @em3dia

For travelers seeking a truly extraordinary culinary experience on the emerald coast of Sardinia, booking a table at ⁠Le Terrazze Ritual in Baja Sardinia is a mandatory addition to your itinerary. Widely celebrated as one of the most visually stunning islands in Italy, Sardinia's dramatic topography is perfectly utilized at this world-class venue. The restaurant and lounge are intricately carved directly into the sheer granite face of an old stone quarry, creating a multi-tiered architectural masterpiece that blends organic rock formations with upscale Mediterranean design. Guests can indulge in high-end, locally sourced seafood dishes while enjoying an unobstructed, panoramic view of the sun setting over the sea from the elevated open-air terraces. Due to its status as one of the most exclusive and romanced dining spots on the island, securing reservations weeks in advance is highly essential during the peak summer season. 🎥 mangiawithmariah | IG