Education & Research Software & Algorithm Provider Human-Machine Interaction Perception & Vision
Interview with Wang Zhongyuan: VLA will survive, but world models are the future.
Original from 36kr.com: 硬氪专访 | 智源研究院院长王仲远:VLA不会死,但世界模型是未来

Interview with Wang Zhongyuan: VLA will survive, but world models are the future.

In recent months, the concept of "World Model" has gained significant traction within the AI and robotics sectors, driven by underlying industry anxieties. As AI technology has rapidly evolved over the past two years, limitations in embodied intelligence have become apparent, revealing that while robots can recognize objects, they struggle to understand physical interactions and causal relationships. The World Model aims to bridge this gap by enabling robots to learn the laws of the physical world.

At the forefront of this exploration is Wang Zhongyuan, the director of the Beijing Academy of Artificial Intelligence, who identifies four distinct paths in the development of World Models. These include language-centered models, pixel-centered models, 3D structure-centered models, and visual representation-centered models. The Beijing Academy is pioneering a fifth approach that integrates language and visual data into a unified latent space representation, allowing for more complex interactions and predictions.

Wang emphasizes that the World Model's potential lies in its ability to enhance embodied intelligence, enabling robots to understand and predict physical interactions over time. He envisions a future where World Models serve as the foundational brain for robots, capable of complex reasoning and decision-making in real-world scenarios. However, he cautions that achieving this goal will require significant advancements in data collection and model training, with a timeline of three to five years anticipated for substantial progress. As the field continues to evolve, the competition will focus on the ability to create models that accurately reflect the complexities of the physical world.

RobotToday Initiative

Robotics needs a service framework.

RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.

Share

Related Articles/News

Microsoft Introduces Provisional Code of Conduct for Future AI Models Amid Industry Concerns

Microsoft Introduces Provisional Code of Conduct for Future AI Models Amid Industry Concerns

Microsoft has unveiled a provisional code of conduct that establishes restrictions for its future artificial intelligence models. This decision follows a growing consensus among AI leaders, including those from Anthropic and OpenAI, advocating for a deceleration in AI development due to rising safety concerns. Mustafa Suleyman, CEO of Microsoft AI, emphas...

Faraday Future Launches Nine New Robot Models to Dominate U.S. Market

Faraday Future Launches Nine New Robot Models to Dominate U.S. Market

Faraday Future (FF) announced the launch of nine new robot models during a press conference on September 19, 2023. These models include humanoid, quadruped, and wheeled-arm robots, marking a significant expansion in FF's product lineup aimed at achieving market dominance in the U.S. robotics sector. This move is notable as it positions FF as the company w...

Robotics Humanoid Robots Automation
Digua Robotics and Giga Vision Collaborate to Integrate World Models into Edge AI Chips

Digua Robotics and Giga Vision Collaborate to Integrate World Models into Edge AI Chips

On September 14, Digua Robotics and Giga Vision announced a strategic collaboration focused on integrating their respective technologies. Giga Vision will provide world models, embodied foundational models, and real-world application experience, while Digua Robotics will contribute an edge AI computing platform, algorithm toolchain, and robotics ecosystem...

Robotics AI Edge Computing
World Models Enhance AI's Transition from Digital to Physical Environments

World Models Enhance AI's Transition from Digital to Physical Environments

As artificial intelligence (AI) evolves from digital interactions to real-world applications, it faces complex challenges in perception, decision-making, and action. World models are seen as a promising solution, enabling AI to predict outcomes and adapt to environmental changes. This technology is crucial for tasks such as autonomous driving and robotics...

World Models AI Development Robotics
Anthropic Disrupts Five Cases of Claude Misuse for Potential Bioweapons Research

Anthropic Disrupts Five Cases of Claude Misuse for Potential Bioweapons Research

Anthropic reported that it intervened in five instances where researchers misused its AI model, Claude, in ways that could have facilitated biological weapons development. These incidents occurred between December 2025 and August 2026, highlighting the challenges of distinguishing legitimate scientific research from potentially harmful activities. The sig...

AI and Robotics
Sixth Graders Program Robots to Breakdance, Highlighting Shift in Robotics Education

Sixth Graders Program Robots to Breakdance, Highlighting Shift in Robotics Education

At Punahou School in Hawaii, a group of sixth graders successfully programmed wheeled robots to perform breakdancing moves. This project involved translating complex physical movements into sequences of commands that machines can understand, encompassing concepts like coordinate transformation and feedback control. With the right educational tools, these ...

Robotics Education Programming Algorithm Development
Anthropic Introduces Text Watermarking for Future Claude Models Amid Regulatory Changes

Anthropic Introduces Text Watermarking for Future Claude Models Amid Regulatory Changes

On August 11, Anthropic announced that its upcoming Claude models will include a text watermark to signify AI-generated content. This initiative aligns with the European Union's AI Act, which mandates watermarks for AI outputs starting August 2, 2026, to combat misleading AI-generated material. While watermarks for images and videos have been effective, t...

Generative-ai Watermark Large-language-models
NOAA Selects Mississippi State University to Lead New Northern Gulf Research Institute

NOAA Selects Mississippi State University to Lead New Northern Gulf Research Institute

NOAA has appointed Mississippi State University to host the Cooperative Institute in the Gulf of America (CINGA), aimed at enhancing research collaboration on Gulf ecosystems. This initiative builds on a 20-year partnership with NOAA, focusing on science-based decision-making for the northern Gulf region. The establishment of CINGA is significant as it co...

noaa mississippi state university northern gulf research institute
Teledyne Webb Research Secures U.S. Navy Contract for APEX Profiling Floats

Teledyne Webb Research Secures U.S. Navy Contract for APEX Profiling Floats

Teledyne Webb Research has been awarded a contract by the U.S. Navy for the provision of APEX profiling floats. This contract highlights the Navy's commitment to enhancing oceanographic research capabilities through advanced technology. The APEX profiling floats are designed to collect critical ocean data, which is essential for various applications, incl...

teledyne webb research contract award u.s. navy
China’s Dexterous Hand Industry, 2026 Part 3 — The Players

China’s Dexterous Hand Industry, 2026 Part 3 — The Players

Five vendor archetypes, MIR’s composite top ten, eight company profiles and six field deployments — every firm linked to the RobotToday supplier directory.

Components and Hardware Humanoids
AGIBOT Open-Sources 11,430 Robot Trajectories to Enhance Reinforcement Learning Research

AGIBOT Open-Sources 11,430 Robot Trajectories to Enhance Reinforcement Learning Research

AGIBOT, a Chinese robotics company, has released its WORLD 2026 dataset, featuring 11,430 robot trajectories aimed at advancing research in reinforcement learning for embodied AI. This dataset emphasizes learning from real-world interactions, capturing a range of experiences including successes, failures, and human interventions. The significance of this ...

AI and Robotics
China’s Dexterous Hand Industry, 2026 Part 1 — The Demand Picture

China’s Dexterous Hand Industry, 2026 Part 1 — The Demand Picture

MIR Databank puts China dexterous hand shipments above 30,000 units in 2026, yet only 3.5% of 2025 demand came from factory production. Part 1 of 3.

Components and Hardware Humanoids

Related Suppliers

BioRobotics Laboratory, Korea University of Technology and Education

South Korean university research laboratory focusing on haptics, exoskeletons, rehabilitation robotics, and autonomous ground vehicles.

Human Engineering Research Laboratories (HERL)

VA–Univ. of Pittsburgh research center founded 1994, developing wheelchair robotics and assistive mobility tech for veterans and people with disabilities.

Education & Research Healthcare & Senior Care Human-Machine Interaction

Robotics Education & Competition Foundation

The REC Foundation is a US nonprofit that runs VEX-based robotics and drone competitions reaching 1.1 million students across 70 countries.

IT+Robotics (IT)

IT+Robotics develops 3D machine vision systems for industrial robotics including EyeT+ random bin-picking, adaptive machining path generation and automated 3D completeness inspection for production quality control.

Software & Algorithm Provider Perception & Vision Control & Software Systems

Magellium (FR)

French software and data analytics company specializing in satellite image processing, remote sensing, and geospatial intelligence. Serves space, defense, and environmental monitoring sectors. Website unavailable at scrape time.

Software & Algorithm Provider Research Institute Perception & Vision

Ortelio Ltd

Ortelio Ltd is a UK cloud robotics research company founded in 2011 that develops AI platforms giving robots intelligence for service and healthcare tasks.

Education & Research Software & Algorithm Provider Research Institute

Human in Motion Robotics

Canadian company developing hands-free lower-limb exoskeletons for spinal cord injury and stroke rehabilitation.

Human-Machine Interaction Collaborative Mobile Manipulator

LG Display

Seoul display maker (KRX:034220, NYSE:LPL), formed 1999, that launched plastic OLED humanoid-robot face panels for industrial settings in 2026.

Roboblock System Co., Ltd.

South Korean robotics distribution and development company founded in 1997; supplies AI robots, drones, humanoid kits, and VR/AR systems internationally.

Robot Manufacturer Software & Algorithm Provider Human-Machine Interaction

Australian Centre for Robotic Vision

World's first research centre specialising in robotic vision, an ARC Centre of Excellence led by QUT.

Perception & Vision Control & Software Systems Research Institute

Machine Vision Lighting Inc.

Tokyo firm whose patented VISA-Method lighting delivers uniform illumination and one-step 3-D imaging for machine-vision inspection.

Tangram Vision

Perception infrastructure software and hardware company enabling reliable sensor calibration and integration for robotics and autonomous systems.

inJoin the RobotToday community on LinkedIn

Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.