AI research - aitoearn

AI conference

Chinese Information Processing Society Academic Annual Conference & 2nd China Large Model Conference (CIPS & CLM 2025) Grand Opening in Beijing on October 28, 2025

China National Conference on Computational Linguistics 2025 & 2nd China Large Model Conference (CIPS & CLM 2025) Open in Beijing On October 28, the China National Conference on Computational Linguistics 2025 and the 2nd China Large Model Conference (CIPS & CLM 2025) officially opened in Beijing, welcoming over 600 attendees.

robotics

A New Model Lets Robots “Imagine” the World — Joint Release by PI Co‑Founders & Tsinghua’s Chen Jianyu Team

CTRL-WORLD: A Controllable Generative World Model for Robot Manipulation In recent days, Chelsea Finn, co‑founder of Physical Intelligence (PI), has been engaging with a new world model research project from her Stanford group. > Generating visually appealing videos is easy — the challenge is building a truly general‑purpose model

AI conference

Chinese Information Processing Society of China 2025 Annual Conference & 2nd China Large Model Summit (CIPS & CLM 2025) Opens Grandly in Beijing on October 28

China Chinese Information Processing Society Academic Annual Conference 2025 & 2nd China Large Language Model Conference (CIPS & CLM 2025) Grand Opening in Beijing, October 28–29, 2025 Over 600 scholars, researchers, and industry leaders gathered in Beijing for the joint opening of the China Chinese Information Processing Society Academic

robotics

A Model That Lets Robots Learn the World Through “Imagination” — Co-created by PI Founding Research Group & Tsinghua’s Chen Jianyu Team

CTRL-WORLD: Controllable Generative World Model for Robotics Background In recent days, Physical Intelligence (PI) co‑founder Chelsea Finn has expressed strong support for a new robotics world model project from Stanford, co‑developed with Chen Jianyu’s team at Tsinghua University. Finn emphasizes: > “It’s easy to generate videos

SVG model

VAE Hit Again! Tsinghua & Kuaishou Unveil SVG Diffusion Model with 6200% Training Boost and 3500% Generation Speedup

SVG: A Latent Diffusion Model Without VAE Right after Xie Saining announced the retirement of VAE from the image generation field, Tsinghua University and Kuaishou’s Kolin team introduced SVG — a latent diffusion model that completely avoids using a VAE. This approach achieves: * 62× improvement in training efficiency * 35× boost

embodied AI

In-Depth | Reaching World No. 1: Global Embodied AI Core Circle Votes With Their Feet, Breakthrough in the Industry’s Data Bottleneck Emerges

Galaxea: The Open-World Dataset Disrupting Embodied Intelligence In the embodied intelligence field—where high-quality data is rare—an open-source dataset from China has shattered expectations. In just two months, it has surpassed 400,000 downloads, becoming a de facto standard among developers worldwide. --- Introduction The shortage of high-quality data

Yoshua Bengio

Yoshua Bengio Becomes the World’s First Scientist with One Million Citations

The World’s First Scholar with Over One Million Citations Yoshua Bengio Defines the AI Era World-renowned computer scientist Yoshua Bengio has just become the first person on Google Scholar to surpass one million citations — a landmark moment in the history of AI research. To illustrate the scale of this

AI research

Father of Transformers “Defects”: I’m Done with Transformers! Warning: AI Research Is Becoming Too Narrow — We Need New Architectures

The AI Field Is More Funded Than Ever — Yet Research Is Narrowing > “I’m really done with Transformers.” — Llion Jones, co‑author of the Transformer and co‑founder of Sakana AI At the TED AI conference in San Francisco, Llion Jones — celebrated as a “founding father of the generative

multimodal AI

Relentless Pursuit of "Text Intelligence": The Next Frontier in Multimodal Research

Everywhere We Look, Text Holds Hidden Challenges — Time to Talk About Multimodal Text Intelligence AI in Medical Diagnosis — A Growing Trend Recently, we reported on a striking phenomenon: more and more people are consulting AI for medical advice, especially when the cause of illness is unclear. Some even claim AI

visual tokens

Zhipu Was Just Unlucky — Visual Token Research Collided with DeepSeek

Glyph & DeepSeek-OCR: The Visual Token Showdown It’s quite the coincidence — Zhipu and DeepSeek have crossed paths again. Just one day after DeepSeek-OCR debuted, Zhipu open-sourced its own visual token framework: Glyph. And on the same stage? Enter Karpathy, who’s been showering DeepSeek with likes the past few

DeepSeek

DeepSeek’s Ultimate Ambition: Transforming the Core Language of Large Models into Images | DeepSeek Paper Analysis

DeepSeek-OCR — Optical Compression for Long-Context AI When we think about DeepSeek, multimodality is rarely the first association. However, on October 20, DeepSeek released DeepSeek-OCR as an open-source project — an OCR (Optical Character Recognition) model achieving SOTA results on benchmarks like OmniDocBench. Why the sudden move into OCR? The answer: the

AI research

All AI Gifts Have Long Been Priced in Secret | Peking University’s Latest Paper Explained

AI Has Lightened Our Workload — But We Need a System for Deep Thinking Source丨Tencent Technology Generative AI is reshaping industries — and the way humans write, think, and perceive. After the launch of ChatGPT 3.5, many believed AI would bring “work equality”. Early studies even suggested it might reduce productivity