Loading video...

Video Failed to Load

Go Home

@geometrynodes tree gen + simulation. Needs some springiness.. kinda sloppy tree #b3d Based on harry blends wiggle stuff.

59,760 views • 4 months ago •via X (Twitter)

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Baobab, a prehistoric species which predates both mankind and splitting of continents over 200 million years ago. Baobabs (Adansonia) are distinctive trees with incredibly large trunks. They can store tremendous amounts of water, as their trunks noticeably swell during the rainy season. Tree is native to African savannah, where climate is extremely dry and arid, it is a symbol of life and positivity in a landscape where little else can thrive. African baobab (A. digitata) has a vast range throughout the arid parts of Africa; six additional species are native to the Indian Ocean island of Madagascar, off south-east Africa; and an eighth species is native to north-west Australia. The largest known living baobab is the Sagole Big Tree, a specimen of A. digitata located in Masisi, Vhembe, South Africa, near the border with Zimbabwe. Based on its most recent measurements, Sagole has an extremely large base that covers 60.6m², a height of 19.8m and a total wood and bark volume of 414m³. Its aboveground dry mass is estimated to be 54 tonnes. Baobab trees grow in 32 African countries. Tree known for its longevity and some specimens in Africa have been dated to between 1100-2500 years old and reach up to 30m high and up to an enormous 50m in circumference. Baobab trees can provide shelter, food and water for animals and humans, many savannah communities made their homes near Baobab trees. Baobab also looms among ancient mounds and remains scattered around them are invariably early medieval or Portuguese. Until 2018, the largest living baobab was sacred Tsitakakoike Baobab, a specimen of the endangered species Adansonia grandidieri, which grew near Andombiro in the Ambiky Forest of south-west Madagascar. The incredibly stout and compact tree had a cylindrical trunk with a base that covered 59.6 m², a height of 14.6m and a total volume of 455 m³ - 380 m³ of which was trunk and 75 m³ of which was canopy. It partially broke and collapsed in February 2018 leaving about 40% of the tree still standing, but this was expected to also collapse soon after. An even larger African baobab tree (A. digitata) alive during 21st Century was the Platland/Sunland Tree of Modjadjiskloof, South Africa, with a base of 67.9 m², height of 18.9m and a total wood and bark volume of 448 m³. Unfortunately, a large portion of the Platland Tree collapsed and died in 2016, leaving the Sagole Big Tree to claim the top spot. Baobabs have among the lightest wood for any tree. Balsa wood is well known to model aeroplane makers as one of the lightest and softest woods, with a wood density that averages around 0.15 g/cm³, yet baobab wood is even lighter, averaging 0.13 g/cm³. As a result, the aboveground dry mass of the Platland baobab was estimated at only 58 tonnes and about 59 tonnes for Tsitakakoike. In terms of mass, giant gum trees (Eucalyptus) of Australia are the largest hardwood trees. The Sagole Big Tree has been carbon-dated to 800 years old, the Platland Tree to 1100 years and Tsitakakoike to 1270 years. 🎥© Africa Live #archaeohistories

Archaeo - Histories

83,444 views • 1 year ago

Researchers built a new RAG approach that: - does not need a vector DB. - does not embed data. - involves no chunking. - performs no similarity search. And it hit 98.7% accuracy on a financial benchmark (SOTA). Here's the core problem with RAG that this new approach solves: Traditional RAG chunks documents, embeds them into vectors, and retrieves based on semantic similarity. But similarity ≠ relevance. When you ask "What were the debt trends in 2023?", a vector search returns chunks that look similar. But the actual answer might be buried in some Appendix, referenced on some page, in a section that shares zero semantic overlap with your query. Traditional RAG would likely never find it. PageIndex (open-source) solves this. Instead of chunking and embedding, PageIndex builds a hierarchical tree structure from your documents, like an intelligent table of contents. Then it uses reasoning to traverse that tree. For instance, the model doesn't ask: "What text looks similar to this query?" Instead, it asks: "Based on this document's structure, where would a human expert look for this answer?" That's a fundamentally different approach with: - No arbitrary chunking that breaks context. - No vector DB infrastructure to maintain. - Traceable retrieval to see exactly why it chose a specific section. - The ability to see in-document references ("see Table 5.3") the way a human would. But here's the deeper issue that it solves. Vector search treats every query as independent. But documents have structure and logic, like sections that reference other sections and context that builds across pages. PageIndex respects that structure instead of flattening it into embeddings. Do note that this approach may not make sense in every use case since traditional vector search is still fast, simple, and works well for many applications. But for professional documents that require domain expertise and multi-step reasoning, this tree-based, reasoning-first approach shines. For instance, PageIndex achieved 98.7% accuracy on FinanceBench, significantly outperforming traditional vector-based RAG systems on complex financial document analysis. Everything is fully open-source, so you can see the full implementation in GitHub and try it yourself. I have shared the GitHub repo in the replies!

Avi Chawla

972,565 views • 6 months ago