#decodingaianddatascience 搜尋結果

未找到 "#decodingaianddatascience" 的結果

🚀 DeepSeek-OCR — the new frontier of OCR from @deepseek_ai , exploring optical context compression for LLMs, is running blazingly fast on vLLM ⚡ (~2500 tokens/s on A100-40G) — powered by vllm==0.8.5 for day-0 model support. 🧠 Compresses visual contexts up to 20× while keeping…

vllm_project's tweet image. 🚀 DeepSeek-OCR — the new frontier of OCR from @deepseek_ai , exploring optical context compression for LLMs, is running blazingly fast on vLLM ⚡ (~2500 tokens/s on A100-40G) — powered by vllm==0.8.5 for day-0 model support.

🧠 Compresses visual contexts up to 20× while keeping…
vllm_project's tweet image. 🚀 DeepSeek-OCR — the new frontier of OCR from @deepseek_ai , exploring optical context compression for LLMs, is running blazingly fast on vLLM ⚡ (~2500 tokens/s on A100-40G) — powered by vllm==0.8.5 for day-0 model support.

🧠 Compresses visual contexts up to 20× while keeping…
vllm_project's tweet image. 🚀 DeepSeek-OCR — the new frontier of OCR from @deepseek_ai , exploring optical context compression for LLMs, is running blazingly fast on vLLM ⚡ (~2500 tokens/s on A100-40G) — powered by vllm==0.8.5 for day-0 model support.

🧠 Compresses visual contexts up to 20× while keeping…

Machine Learning Image Credit : DataInterview

PythonPr's tweet image. Machine Learning
Image Credit : DataInterview

This is the JPEG moment for AI. Optical compression doesn't just make context cheaper. It makes AI memory architectures viable. Training data bottlenecks? Solved. - 200k pages/day on ONE GPU - 33M pages/day on 20 nodes - Every multimodal model is data-constrained. Not anymore.…

RayFernando1337's tweet image. This is the JPEG moment for AI.

Optical compression doesn't just make context cheaper. It makes AI memory architectures viable.

Training data bottlenecks? Solved.
- 200k pages/day on ONE GPU
- 33M pages/day on 20 nodes
- Every multimodal model is data-constrained. Not anymore.…

DeepSeek-OCR looks impressive, but its core idea is not new. Input “Text” as “Image” — already explored by: LANGUAGE MODELING WITH PIXELS (Phillip et al., ICLR 2023) CLIPPO: Image-and-Language Understanding from Pixels Only (Michael et al. CVPR 2023) Pix2Struct: Screenshot…

awinyimgprocess's tweet image. DeepSeek-OCR looks impressive, but its core idea is not new.

Input “Text” as “Image” — already explored by:
LANGUAGE MODELING WITH PIXELS (Phillip et al., ICLR 2023)
CLIPPO: Image-and-Language Understanding from Pixels Only (Michael et al. CVPR 2023)
Pix2Struct: Screenshot…
awinyimgprocess's tweet image. DeepSeek-OCR looks impressive, but its core idea is not new.

Input “Text” as “Image” — already explored by:
LANGUAGE MODELING WITH PIXELS (Phillip et al., ICLR 2023)
CLIPPO: Image-and-Language Understanding from Pixels Only (Michael et al. CVPR 2023)
Pix2Struct: Screenshot…
awinyimgprocess's tweet image. DeepSeek-OCR looks impressive, but its core idea is not new.

Input “Text” as “Image” — already explored by:
LANGUAGE MODELING WITH PIXELS (Phillip et al., ICLR 2023)
CLIPPO: Image-and-Language Understanding from Pixels Only (Michael et al. CVPR 2023)
Pix2Struct: Screenshot…
awinyimgprocess's tweet image. DeepSeek-OCR looks impressive, but its core idea is not new.

Input “Text” as “Image” — already explored by:
LANGUAGE MODELING WITH PIXELS (Phillip et al., ICLR 2023)
CLIPPO: Image-and-Language Understanding from Pixels Only (Michael et al. CVPR 2023)
Pix2Struct: Screenshot…

🚨 DeepSeek just did something wild. They built an OCR system that compresses long text into vision tokens literally turning paragraphs into pixels. Their model, DeepSeek-OCR, achieves 97% decoding precision at 10× compression and still manages 60% accuracy even at 20×. That…

godofprompt's tweet image. 🚨 DeepSeek just did something wild.

They built an OCR system that compresses long text into vision tokens  literally turning paragraphs into pixels.

Their model, DeepSeek-OCR, achieves 97% decoding precision at 10× compression and still manages 60% accuracy even at 20×. That…

BOOOOOOOM! CHINA DEEPSEEK DOES IT AGAIN! An entire encyclopedia compressed into a single, high-resolution image! — A mind-blowing breakthrough. DeepSeek-OCR, unleashed an electrifying 3-billion-parameter vision-language model that obliterates the boundaries between text and…

BrianRoemmele's tweet image. BOOOOOOOM!

CHINA DEEPSEEK DOES IT AGAIN!

An entire encyclopedia compressed into a single, high-resolution image!

—

A mind-blowing breakthrough. DeepSeek-OCR, unleashed an electrifying 3-billion-parameter vision-language model that obliterates the boundaries between text and…

A more serious thread on the DeepSeek-OCR hype / serious misinterpretation going on. 1. On token reduction via representing text in images, researchers from Cambridge have previously shown that 500x prompt token compression is possible (ACL'25, Li, Su, and Collier). Without…

Kangwook_Lee's tweet image. A more serious thread on the DeepSeek-OCR hype / serious misinterpretation going on.

1. 
On token reduction via representing text in images, researchers from Cambridge have previously shown that 500x prompt token compression is possible (ACL'25, Li, Su, and Collier). 

Without…
Kangwook_Lee's tweet image. A more serious thread on the DeepSeek-OCR hype / serious misinterpretation going on.

1. 
On token reduction via representing text in images, researchers from Cambridge have previously shown that 500x prompt token compression is possible (ACL'25, Li, Su, and Collier). 

Without…
Kangwook_Lee's tweet image. A more serious thread on the DeepSeek-OCR hype / serious misinterpretation going on.

1. 
On token reduction via representing text in images, researchers from Cambridge have previously shown that 500x prompt token compression is possible (ACL'25, Li, Su, and Collier). 

Without…

👨‍🔧 Inside the smart design of DeepSeek OCR DeepSeek-OCR looks like just another OCR model at first glance, something that reads text from images. But it’s not just that. What they really built is a new way for AI models to store and handle information. Normally, when AI reads…

rohanpaul_ai's tweet image. 👨‍🔧 Inside the smart design of DeepSeek OCR

DeepSeek-OCR looks like just another OCR model at first glance, something that reads text from images. But it’s not just that. 

What they really built is a new way for AI models to store and handle information.

Normally, when AI reads…

DeepSeek-OCR Contexts Optical Compression

_akhaliq's tweet image. DeepSeek-OCR

Contexts Optical Compression

🤖 Smarter generation for smarter creators. #ImagenAI uses DALL·E & Stable Diffusion to ensure consistent, reliable, and high-quality image creation. 🌍 imagen.network

Imagen_Network's tweet image. 🤖 Smarter generation for smarter creators.  
#ImagenAI uses DALL·E & Stable Diffusion to ensure consistent, reliable, and high-quality image creation.  
🌍 imagen.network

DeepSeek-OCR just dropped. 🔥 Sets a new standard for open-source OCR A 3B-parameter vision-language model designed for high-performance optical character recognition and structured document conversion. - Can parse and re-render charts in HTML - Optical Context Compression:…

rohanpaul_ai's tweet image. DeepSeek-OCR just dropped. 🔥

Sets a new standard for open-source OCR

A 3B-parameter vision-language model designed for high-performance optical character recognition and structured document conversion. 

-  Can parse and re-render charts in HTML

- Optical Context Compression:…

Defrag: Signals DEFRAG: Signals is a co-created collection where the community defined the graphic layers to forge their own 1/1sbased on my initial artwork. opensea.io/collection/def…

costajpeg's tweet image. Defrag: Signals

DEFRAG: Signals is a co-created collection where the community defined the graphic layers to forge their own 1/1sbased on my initial artwork.

opensea.io/collection/def…
costajpeg's tweet image. Defrag: Signals

DEFRAG: Signals is a co-created collection where the community defined the graphic layers to forge their own 1/1sbased on my initial artwork.

opensea.io/collection/def…
costajpeg's tweet image. Defrag: Signals

DEFRAG: Signals is a co-created collection where the community defined the graphic layers to forge their own 1/1sbased on my initial artwork.

opensea.io/collection/def…
costajpeg's tweet image. Defrag: Signals

DEFRAG: Signals is a co-created collection where the community defined the graphic layers to forge their own 1/1sbased on my initial artwork.

opensea.io/collection/def…

IT FREAKING WORKED! At 4am today I just proved DeepSeek-OCR AI can scan an entire microfiche sheet and not just cells and retain 100% of the data in seconds… AND Have a full understanding of the text/complex drawings and their context. I just changed offline data curation!

BrianRoemmele's tweet image. IT FREAKING WORKED!

At 4am today I just proved DeepSeek-OCR AI can scan an entire microfiche sheet and not just cells and retain 100% of the data in seconds…

AND

Have a full understanding of the text/complex drawings and their context.

I just changed offline data curation!

BOOOOOOOM! CHINA DEEPSEEK DOES IT AGAIN! An entire encyclopedia compressed into a single, high-resolution image! — A mind-blowing breakthrough. DeepSeek-OCR, unleashed an electrifying 3-billion-parameter vision-language model that obliterates the boundaries between text and…

BrianRoemmele's tweet image. BOOOOOOOM!

CHINA DEEPSEEK DOES IT AGAIN!

An entire encyclopedia compressed into a single, high-resolution image!

—

A mind-blowing breakthrough. DeepSeek-OCR, unleashed an electrifying 3-billion-parameter vision-language model that obliterates the boundaries between text and…


Unlike closed AI labs, DeepSeek proves they are truly open research Their OCR paper treats paragraphs as pixels and is 60x leap more efficient than traditional LLMs Small super efficient models are the future

bindureddy's tweet image. Unlike closed AI labs, DeepSeek proves they are truly open research 

Their OCR paper treats paragraphs as pixels and is  60x leap more efficient than traditional LLMs 

Small super efficient models are the future

🚀Tired of fragmented VLA codebases? Meet Dexbotic by @Dexmal_AI. An open-source PyTorch-based toolbox that unifies training & evaluation of SOTA Vision-Language-Action models. Go from idea to result faster.

Jeffar_AI's tweet image. 🚀Tired of fragmented VLA codebases? 

Meet Dexbotic by @Dexmal_AI. An open-source 
PyTorch-based toolbox that unifies training & 

evaluation of SOTA Vision-Language-Action models. Go from idea to result faster.

Very powerful machine to transform your dataset.

Eyowhite3's tweet image. Very powerful machine to transform your dataset.

OSI Model data encapsulation and de-encapsulation

thatstraw's tweet image. OSI Model data encapsulation and de-encapsulation

「DecopyAI」で画像がAI生成か否かを判定させてみる。写真のフィルタ加工で作った画像をAI呼ばわりされたりして結構デタラメ判定だが、それでも最近の中では比較的マシな的中率かもしれない。 画像はAIで線画を生成してコピックでアナログ着色をし、それを撮った写真の判定。

monaka_k's tweet image. 「DecopyAI」で画像がAI生成か否かを判定させてみる。写真のフィルタ加工で作った画像をAI呼ばわりされたりして結構デタラメ判定だが、それでも最近の中では比較的マシな的中率かもしれない。

画像はAIで線画を生成してコピックでアナログ着色をし、それを撮った写真の判定。

DeepSeek-OCR is out! 🔥 my take ⤵️ > pretty insane it can parse and re-render charts in HTML > it uses CLIP and SAM features concatenated, so better grounding > very efficient per vision tokens/performance ratio > covers 100 languages

mervenoyann's tweet image. DeepSeek-OCR is out! 🔥 my take ⤵️ 
> pretty insane it can parse and re-render charts in HTML
> it uses CLIP and SAM features concatenated, so better grounding
> very efficient per vision tokens/performance ratio
> covers 100 languages

Loading...

Something went wrong.


Something went wrong.


United States Trends