fzliu

30 ms·

571.
Matryoshka Representation Learning(arxiv.org)
83 points by fzliu 3y ago|11 comments
572.
Building a Video Games Recommender System(blog.stackademic.com)
1 points by fzliu 3y ago|0 comments
573.
Exploring the Intel AVX-512 Integration with the Milvus Vector Database(community.intel.com)
2 points by fzliu 3y ago|0 comments
574.
Scalable AI Feature Store with Milvus on Kubernetes(pulumi.com)
1 points by fzliu 3y ago|0 comments
575.
Things to Know About Python Super, 1 of 3 (2008)(artima.com)
1 points by fzliu 3y ago|0 comments
576.
Entering the Era of Intelligent Search(bvp.com)
1 points by fzliu 3y ago|0 comments
577.
What killed the fight scene?(fastcompany.com)
20 points by fzliu 3y ago|57 comments
578.
Locally-Adaptive Vector Quantization(arxiv.org)
2 points by fzliu 3y ago|0 comments
579.
Intel AMX(intel.com)
21 points by fzliu 3y ago|5 comments
580.
Embedding models saturate at 300-500 tokens(twitter.com)
2 points by fzliu 3y ago|0 comments
581.
by fzliu 3y ago
"AGI" will never be achieved without building a model that a) _continually_ learns, and b) learns from not just text, but from combined auditory and visual (multimodal) sensory information as well. The reason a 16-year-old can lea
582.
by fzliu 3y ago
In my mind, what's more crucial here is code for downloading/scraping and labeling the data, not the model architecture nor training script. As much as I appreciate Mis(x)tral, I would've loved it even more if they released c
583.
by fzliu 3y ago
You need different indexing algorithms for different use cases - brute-force indexing, for example, is "SOTA" when it comes to recall (100%). If you have multiple use cases or if you might have domain shift, you'll want a vec
584.
The Musk Algorithm(world.hey.com)
2 points by fzliu 3y ago|0 comments
585.
by fzliu 3y ago
Throwing a few more on here (mix of beginner and advanced): - Wikipedia article: https://en.wikipedia.org/wiki/Vector_database - Vector Database 101: https://zilliz.com/learn/introduction-to-unstru
586.
by fzliu 3y ago
Like IVF, Annoy partitions the entire embedding space into high-dimensional polygons. The difference is how the two algorithms do it - IVF ( https://zilliz.com/learn/vector-index ) uses centroids, while Annoy ( https:&#x
587.
Professor Forcing(github.com)
1 points by fzliu 3y ago|0 comments
588.
PyTorch for ROCm(rocm.docs.amd.com)
10 points by fzliu 3y ago|0 comments
589.
The Pocket Guide of Essential Early-Stage Sales Advice(jjellyfish.com)
2 points by fzliu 3y ago|0 comments
590.
Open Source Toolkit for Cheminformatics(rdkit.org)
1 points by fzliu 3y ago|0 comments
591.
Tokens, N-Grams, and Bag-of-Words Models(zilliz.com)
1 points by fzliu 3y ago|0 comments
592.
Tokens, N-Grams, and Bag-of-Words Models(zilliz.com)
4 points by fzliu 3y ago|0 comments
593.
Tensor Puzzles(github.com)
1 points by fzliu 3y ago|0 comments
594.
by fzliu 3y ago
Most breakthroughs are discovered accidentally and retroactively, so I'd think that having multiple breakthrough papers is fairly uncommon.
595.
Why America Still Doesn't Have Fast Trains(time.com)
8 points by fzliu 3y ago|0 comments
596.
The most important announcement from re:Invent: S3 Express(scalevp.com)
2 points by fzliu 3y ago|0 comments
597.
Vector databases: What makes each one different?(thedataquarry.com)
3 points by fzliu 3y ago|0 comments
598.
Sequential Modeling Enables Scalable Learning for Large Vision Models(arxiv.org)
3 points by fzliu 3y ago|0 comments
599.
Progression of DuckDB Performance(pdeyhim.medium.com)
1 points by fzliu 3y ago|0 comments
600.
Sembly: Collaborative Braintrust(sembly.com)
1 points by fzliu 3y ago|0 comments

More ›