Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Given an embedding vector, you can tell which model produced it. I trained a 0.8M transformer that fingerprints embedding models by reading raw float digits (vocab size: 15). Full end-to-end, zero feature engineering.

Han Xiao

17,686 subscribers

19,688 Aufrufe • vor 3 Monaten •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

RTFM can be seen as a learned renderer: it is an autoregressive diffusion transformer trained end-to-end on large-scale video data, and it learns to model 3D geometry, reflections, shadows and more just by observing them in its training set.

RTFM can be seen as a learned renderer: it is an autoregressive diffusion transformer trained end-to-end on large-scale video data, and it learns to model 3D geometry, reflections, shadows and more just by observing them in its training set.

World Labs

15,870 Aufrufe • vor 8 Monaten

Announcing CT Foundation, a new medical imaging embedding tool that accepts a computed tomography (CT) volume as input and returns a small, information-rich numerical embedding that can be used to rapidly train models. Learn more and try it out yourself →

Announcing CT Foundation, a new medical imaging embedding tool that accepts a computed tomography (CT) volume as input and returns a small, information-rich numerical embedding that can be used to rapidly train models. Learn more and try it out yourself →

Google AI

246,314 Aufrufe • vor 1 Jahr

Embedding features learned with sparse autoencoders can make semantic edits to text ✨ (+ a reading/highlighting demo) I've built an interface to explore and visualize GPT-4 labelled features learned from a text embedding model's latent space. Here's a little video, more in 👇

Embedding features learned with sparse autoencoders can make semantic edits to text ✨ (+ a reading/highlighting demo) I've built an interface to explore and visualize GPT-4 labelled features learned from a text embedding model's latent space. Here's a little video, more in 👇

Linus

51,614 Aufrufe • vor 2 Jahren

Most don't know (1) how easy it is to invert embedding vectors back into sentences, (2) this is a perfect task text diffusion models. Here's a 78M parameter model and live demo that recovers 80% of tokens from Qwen3-Embedding and EmbeddingGemma vectors. Works even on multilingual input.

Most don't know (1) how easy it is to invert embedding vectors back into sentences, (2) this is a perfect task text diffusion models. Here's a 78M parameter model and live demo that recovers 80% of tokens from Qwen3-Embedding and EmbeddingGemma vectors. Works even on multilingual input.

Jina AI

12,813 Aufrufe • vor 4 Monaten

i can now search my entire ad library by concept. "ugc testimonials with urgency hooks" and it just finds them. here's how i built it: searchable ad swipe file > drag in creatives > AI analyzes them > search by concept > find anything in seconds -> uses google's new multimodal embedding model. images, video, and text all searchable together. -> type "ugc testimonials with urgency hooks" and it finds them. no manual tags needed. -> every ad gets auto-analyzed by gemini on import. hook type, structure, audience, CTA. all searchable. -> dual-vector scoring. visual embedding + text embedding. takes the stronger signal. full build breakdown in video + notion guide in the comments

i can now search my entire ad library by concept. "ugc testimonials with urgency hooks" and it just finds them. here's how i built it: searchable ad swipe file > drag in creatives > AI analyzes them > search by concept > find anything in seconds -> uses google's new multimodal embedding model. images, video, and text all searchable together. -> type "ugc testimonials with urgency hooks" and it finds them. no manual tags needed. -> every ad gets auto-analyzed by gemini on import. hook type, structure, audience, CTA. all searchable. -> dual-vector scoring. visual embedding + text embedding. takes the stronger signal. full build breakdown in video + notion guide in the comments

J.B.

27,000 Aufrufe • vor 2 Monaten

[1/n] Do distinct large models admit a simple map that aligns their embedding spaces? We show that across multimodal contrastive models—trained on different data and architectures—an orthogonal map aligns image embeddings. Strikingly, the same map also aligns text embeddings.

[1/n] Do distinct large models admit a simple map that aligns their embedding spaces? We show that across multimodal contrastive models—trained on different data and architectures—an orthogonal map aligns image embeddings. Strikingly, the same map also aligns text embeddings.

Sharut Gupta

36,956 Aufrufe • vor 3 Monaten

I calculated a poster-size Transformer model by hand✍️

I calculated a poster-size Transformer model by hand✍️

Tom Yeh

112,781 Aufrufe • vor 1 Jahr

New short course on advanced retrieval for RAG (retrieval augmented generation)! RAG fetches relevant documents to give context to an LLM. In Advanced Retrieval for AI with Chroma, taught by Chroma founder anton 🇺🇸, you’ll learn: (i) Query expansion using an LLM to rewrite and improve a query, by either generating either additional relevant queries or a hypothetical answer to the query. (ii) Reranking using a cross-encoder - a model trained to measure similarity between two inputs presented simultaneously. Reranking reorders retrieved documents based on the cross-encoder similarity measure. (iii) Constructing and training an Embedding Adaptor, which is a model that adapts the embedding values to be more relevant to your use case. Each of these techniques can help you build much better RAG systems. Please sign up for the course here:

New short course on advanced retrieval for RAG (retrieval augmented generation)! RAG fetches relevant documents to give context to an LLM. In Advanced Retrieval for AI with Chroma, taught by Chroma founder anton 🇺🇸, you’ll learn: (i) Query expansion using an LLM to rewrite and improve a query, by either generating either additional relevant queries or a hypothetical answer to the query. (ii) Reranking using a cross-encoder - a model trained to measure similarity between two inputs presented simultaneously. Reranking reorders retrieved documents based on the cross-encoder similarity measure. (iii) Constructing and training an Embedding Adaptor, which is a model that adapts the embedding values to be more relevant to your use case. Each of these techniques can help you build much better RAG systems. Please sign up for the course here:

Andrew Ng

191,219 Aufrufe • vor 2 Jahren

1. Project Digits: NVIDIA's Latest AI Supercomputer Introduces an exciting breakthrough with the new NVIDIA GB10 Superchip, offering a compact AI supercomputer that efficiently handles 200B-parameter models. NVIDIA’s Project DIGITS is a powerhouse you can hold in the palm of your hand. With two Project DIGITS units, developers can easily work with AI models up to 405B parameters in size. Available by May.

1. Project Digits: NVIDIA's Latest AI Supercomputer Introduces an exciting breakthrough with the new NVIDIA GB10 Superchip, offering a compact AI supercomputer that efficiently handles 200B-parameter models. NVIDIA’s Project DIGITS is a powerhouse you can hold in the palm of your hand. With two Project DIGITS units, developers can easily work with AI models up to 405B parameters in size. Available by May.

Shruti

236,540 Aufrufe • vor 1 Jahr

Embedding a tracker into an image. Be careful when you open images. RT to help others!!!

Embedding a tracker into an image. Be careful when you open images. RT to help others!!!

Winston Ighodaro

97,972 Aufrufe • vor 6 Monaten

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

New short course: Building Multimodal Search and RAG", by Weaviate AI Database's Sebastia(N_) Witalec ✊🏽✊🏾✊🏿. Contrastive learning is used to train models to map vectors into an embedding space by pulling similar concepts closer together and pushing dissimilar concepts away from each other. This technique is also used to train multimodal embedding models that capture semantic similarity across different modalities like text, images, and audio. These multimodal embeddings can be used to build multimodal search and RAG systems. In this course, you'll learn how contrastive learning works, and how to add multimodality to RAG – so your models can draw on diverse, relevant context to answer questions. For example, a query about a financial report might synthesize information from text snippets, graphs, tables, and slides. You will also learn how visual instruction tuning lets you integrate image understanding into language models, and build a multi-vector recommender system using Weaviate’s open-source vector database. Please sign up here:

Andrew Ng

104,371 Aufrufe • vor 2 Jahren

Her feet can arouse you to no end or crush you like a bug at a whim Model: size 10 🪩

Sensitive content

Her feet can arouse you to no end or crush you like a bug at a whim Model: size 10 🪩

Megashroom Studios 🍄

19,509 Aufrufe • vor 2 Jahren

At Generalist, we’re working towards a future where robots can do anything. To that end, the robots build now, too. We’ve trained a robot to do one-shot assembly, constructing Legos end-to-end: no custom engineering, just pixels in → Lego copies out.

At Generalist, we’re working towards a future where robots can do anything. To that end, the robots build now, too. We’ve trained a robot to do one-shot assembly, constructing Legos end-to-end: no custom engineering, just pixels in → Lego copies out.

Generalist

102,938 Aufrufe • vor 8 Monaten

Q: "Putin mentioned that the war will not end without addressing the root causes of this war. Do you have an understanding what that means?" Trump: "The war's going to end. When it ends, I can't tell you....This gentleman wants it to end and Vladimir Putin wants it to end."

Q: "Putin mentioned that the war will not end without addressing the root causes of this war. Do you have an understanding what that means?" Trump: "The war's going to end. When it ends, I can't tell you....This gentleman wants it to end and Vladimir Putin wants it to end."

The Bulwark

102,638 Aufrufe • vor 10 Monaten

Let's go! 😍 Qwen just released Qwen3-Embedding, a new series of embedding models: 🏆 SOTA performance on MMTEB, MTEB, and MTEB-Code 📏 Three different sizes (0.6B / 4B / 8B) 🌍 Multilingual (119 languages) 💻 Can run in-browser w/ Transformers.js (+ WebGPU acceleration)

Let's go! 😍 Qwen just released Qwen3-Embedding, a new series of embedding models: 🏆 SOTA performance on MMTEB, MTEB, and MTEB-Code 📏 Three different sizes (0.6B / 4B / 8B) 🌍 Multilingual (119 languages) 💻 Can run in-browser w/ Transformers.js (+ WebGPU acceleration)

Xenova

47,143 Aufrufe • vor 1 Jahr

Learn a development pattern to systematically improve the accuracy and reliability of LLM applications in our new short course, Improving Accuracy of LLM Applications, built in partnership with Lamini and Meta, and taught by Lamini’s CEO Sharon Zhou, and Meta’s Senior Director of Partner Engineering, Amit Sangani. (Disclosure: I am an investor in Lamini.) The path to tuning an LLM application can be complex. In this course, you'll learn a systematic sequence of steps for improving accuracy by reducing hallucinations: - Create an evaluation dataset to measure model accuracy - Add prompt engineering and self-reflection - Fine-tune your model including "memory-tuning" which is a new method of embedding facts in an LLM Using the Llama 3-8B parameter model, you will: - Build a text-to-SQL agent with a custom schema and simulate situations where it hallucinates - Understand the difference between instruction fine-tuning, which gives pre-trained LLMs instructions to follow, and memory fine-tuning - See how Performance-Efficient Fine-tuning (PEFT) techniques like Low-Rank Adaptation (LoRA) reduce training time by 100x and Mixture of Memory Experts (MoME) reduces it even further I appreciate Meta releasing the Llama's family of open models -- this course gives an example of the unique type of work that developers can do with such models. Please sign up here:

Learn a development pattern to systematically improve the accuracy and reliability of LLM applications in our new short course, Improving Accuracy of LLM Applications, built in partnership with Lamini and Meta, and taught by Lamini’s CEO Sharon Zhou, and Meta’s Senior Director of Partner Engineering, Amit Sangani. (Disclosure: I am an investor in Lamini.) The path to tuning an LLM application can be complex. In this course, you'll learn a systematic sequence of steps for improving accuracy by reducing hallucinations: - Create an evaluation dataset to measure model accuracy - Add prompt engineering and self-reflection - Fine-tune your model including "memory-tuning" which is a new method of embedding facts in an LLM Using the Llama 3-8B parameter model, you will: - Build a text-to-SQL agent with a custom schema and simulate situations where it hallucinates - Understand the difference between instruction fine-tuning, which gives pre-trained LLMs instructions to follow, and memory fine-tuning - See how Performance-Efficient Fine-tuning (PEFT) techniques like Low-Rank Adaptation (LoRA) reduce training time by 100x and Mixture of Memory Experts (MoME) reduces it even further I appreciate Meta releasing the Llama's family of open models -- this course gives an example of the unique type of work that developers can do with such models. Please sign up here:

Andrew Ng

66,407 Aufrufe • vor 1 Jahr

Google just proved that bigger isn't always better. Their 308M parameter model is outperforming models 2x its size. Google just released 𝗘𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴𝗚𝗲𝗺𝗺𝗮, and it's proving that lightweight embedding models can punch way above their weight class. At just 308M parameters (578MB), it's the new state-of-the-art for models under 500M parameters across MTEB multilingual, English, and code benchmarks. But the really impressive part is that it ranks 8th overall on MTEB(Multilingual, v2) - that's 𝟭𝟳 𝗽𝗹𝗮𝗰𝗲𝘀 above the second-best sub-500M model, and it's delivering performance 𝗰𝗼𝗺𝗽𝗮𝗿𝗮𝗯𝗹𝗲 𝘁𝗼 𝗺𝗼𝗱𝗲𝗹𝘀 𝗻𝗲𝗮𝗿𝗹𝘆 𝗱𝗼𝘂𝗯𝗹𝗲 𝗶𝘁𝘀 𝘀𝗶𝘇𝗲. There are three key parts of their training recipe that sets it apart: 𝟭. 𝗘𝗻𝗰𝗼𝗱𝗲𝗿-𝗗𝗲𝗰𝗼𝗱𝗲𝗿 𝗜𝗻𝗶𝘁𝗶𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 Instead of starting from a decoder-only Gemma 3 model, they first adapted it to encoder-decoder, then used just the encoder. By basing EmbeddingGemma off an LLM that already has world and language understanding, it gives it a stronger starting point. 𝟮. 𝗧𝗵𝗿𝗲𝗲-𝗟𝗼𝘀𝘀 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 They combine three different loss functions, instead of just having one: • Contrastive loss (NCE) with in-batch negatives and hardness weighting • Spread-out regularization to ensure embeddings utilize the full space (for quantization and ANN retrieval) • Embedding matching distillation from Gemini Embedding - not just learning from relevance scores, but directly aligning the embedding space with the teacher model 𝟯. 𝗠𝗼𝗱𝗲𝗹 𝗦𝗼𝘂𝗽𝗶𝗻𝗴 Rather than just averaging checkpoints from the same training run, they use optimization techniques to find multiple specialized training mixtures. Each mixture creates an "expert" model in different domains, and averaging all their parameters creates a final model that's actually better than individual models. Extras: • Matryoshka embeddings supporting 768, 512, 256, and 128 dimensions • Quantization-aware training - maintains quality even at int4 precision • 100+ languages from Gemma 3 pretraining • Exceptional performance on low-resource languages (check their XTREME-UP results) Is it the absolute best embedding model? No - Gemini Embedding still leads overall. But that's not really the point. EmbeddingGemma proves you can achieve state-of-the-art performance in a small package that's actually deployable on-device, in low-latency applications, and in resource-constrained environments. This makes good embeddings accessible for use cases that I'm seeing more and more: offline applications, privacy-sensitive deployments, and high-throughput scenarios where inference cost actually matters. Full paper: Shoutout to the EmbeddingGemma team at Google DeepMind for this awesome open source work 💙 and to Daniel Williams for helping me with this video! 🫶

Google just proved that bigger isn't always better. Their 308M parameter model is outperforming models 2x its size. Google just released 𝗘𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴𝗚𝗲𝗺𝗺𝗮, and it's proving that lightweight embedding models can punch way above their weight class. At just 308M parameters (578MB), it's the new state-of-the-art for models under 500M parameters across MTEB multilingual, English, and code benchmarks. But the really impressive part is that it ranks 8th overall on MTEB(Multilingual, v2) - that's 𝟭𝟳 𝗽𝗹𝗮𝗰𝗲𝘀 above the second-best sub-500M model, and it's delivering performance 𝗰𝗼𝗺𝗽𝗮𝗿𝗮𝗯𝗹𝗲 𝘁𝗼 𝗺𝗼𝗱𝗲𝗹𝘀 𝗻𝗲𝗮𝗿𝗹𝘆 𝗱𝗼𝘂𝗯𝗹𝗲 𝗶𝘁𝘀 𝘀𝗶𝘇𝗲. There are three key parts of their training recipe that sets it apart: 𝟭. 𝗘𝗻𝗰𝗼𝗱𝗲𝗿-𝗗𝗲𝗰𝗼𝗱𝗲𝗿 𝗜𝗻𝗶𝘁𝗶𝗮𝗹𝗶𝘇𝗮𝘁𝗶𝗼𝗻 Instead of starting from a decoder-only Gemma 3 model, they first adapted it to encoder-decoder, then used just the encoder. By basing EmbeddingGemma off an LLM that already has world and language understanding, it gives it a stronger starting point. 𝟮. 𝗧𝗵𝗿𝗲𝗲-𝗟𝗼𝘀𝘀 𝗧𝗿𝗮𝗶𝗻𝗶𝗻𝗴 They combine three different loss functions, instead of just having one: • Contrastive loss (NCE) with in-batch negatives and hardness weighting • Spread-out regularization to ensure embeddings utilize the full space (for quantization and ANN retrieval) • Embedding matching distillation from Gemini Embedding - not just learning from relevance scores, but directly aligning the embedding space with the teacher model 𝟯. 𝗠𝗼𝗱𝗲𝗹 𝗦𝗼𝘂𝗽𝗶𝗻𝗴 Rather than just averaging checkpoints from the same training run, they use optimization techniques to find multiple specialized training mixtures. Each mixture creates an "expert" model in different domains, and averaging all their parameters creates a final model that's actually better than individual models. Extras: • Matryoshka embeddings supporting 768, 512, 256, and 128 dimensions • Quantization-aware training - maintains quality even at int4 precision • 100+ languages from Gemma 3 pretraining • Exceptional performance on low-resource languages (check their XTREME-UP results) Is it the absolute best embedding model? No - Gemini Embedding still leads overall. But that's not really the point. EmbeddingGemma proves you can achieve state-of-the-art performance in a small package that's actually deployable on-device, in low-latency applications, and in resource-constrained environments. This makes good embeddings accessible for use cases that I'm seeing more and more: offline applications, privacy-sensitive deployments, and high-throughput scenarios where inference cost actually matters. Full paper: Shoutout to the EmbeddingGemma team at Google DeepMind for this awesome open source work 💙 and to Daniel Williams for helping me with this video! 🫶

Victoria Slocum

21,211 Aufrufe • vor 7 Monaten

Transformer by hand ✍️ in Excel ~ I just released my first-ever "Full-Stack" implementation of the Transformer model. 👇Download xlsx to give it a try!

Transformer by hand ✍️ in Excel ~ I just released my first-ever "Full-Stack" implementation of the Transformer model. 👇Download xlsx to give it a try!

Tom Yeh

2,997,355 Aufrufe • vor 1 Jahr

After Gothalion called me out for Falsely Embedding and "Scum" today, a Discord member put a video together to sum up the 'drama'🤣 Worth a watch til' the end.

After Gothalion called me out for Falsely Embedding and "Scum" today, a Discord member put a video together to sum up the 'drama'🤣 Worth a watch til' the end.

DannehTV

25,880 Aufrufe • vor 1 Jahr