The Catastrophic Forgetting Problem Nobody Wants to Talk About
New research shows robot AI models forget old skills when learning new ones, and I've seen this exact problem tank entire industries before.
画像クレジット: Lottie animation by Centre Robotics (LottieFiles Free, used with credit). · source
I've seen this movie before. Back in the late 2000s, everyone was convinced we were two years away from self-driving cars because the demos looked so good. The cars could navigate a desert course! They could merge onto highways! What nobody talked about was what happened when you tried to make them do both at the same time, or what happened when conditions changed even slightly from the training data. We're watching the same script play out with robot foundation models, and a batch of new research papers this week makes the problem painfully clear.
The headline finding comes from a team studying vision-language-action models, those supposedly general-purpose robot brains that companies keep promising will make robots as flexible as humans. According to arXiv, when these VLA models try to learn new skills from real-world data, they suffer what the researchers call "significant catastrophic forgetting." In plain English: teach a robot to fold towels and it forgets how to pick up boxes. This isn't a minor tuning problem. This is a fundamental limitation that makes the entire premise of general-purpose robots look shaky.
The researchers built a real-world dataset with four sequential manipulation tasks (rigid object pick-and-place, contact-rich pressing, deformable object folding) and found that current approaches just can't handle learning them in sequence without losing previously learned behaviors. They tested experience replay, which is basically making the robot practice old skills while learning new ones, and found it helps but isn't a magic bullet. The paper's contribution is showing exactly which implementation factors matter, which is useful, but call me old-fashioned: I think we're still treating symptoms instead of the disease.
The Simulation Escape Hatch
The industry's response to the real-world data problem has been predictable: retreat to simulation. A new paper introduces something called GE-Sim 2.0, a "closed-loop video world simulator" that promises to let robots learn from simulated rollouts instead of expensive real-world demonstrations. According to arXiv, the system was trained on thousands of hours of real robot data and can generate a 25-frame rollout in 2.3 seconds on a single H100 GPU. The researchers claim policies trained against these simulated rollouts "translate into measurable real-world gains."
I want to believe this. I really do. The economics of robot learning basically require simulation to work, because you can't have humans teleoperating robots for millions of hours to generate training data. But "measurable real-world gains" is doing a lot of heavy lifting in that sentence. The paper doesn't say how much gain, or under what conditions, or how the gains hold up when the real world throws something unexpected at the robot. These details matter! The sim-to-real gap has been the graveyard of robotics promises for decades.
出典
- Can VLA Models Learn from Real-World Data Continually without Forgetting?· arXiv — cs.RO (Robotics)
- GE-Sim 2.0: A Roadmap Towards Comprehensive Closed-loop Video World Simulators for Robotic Manipulation· arXiv — cs.RO (Robotics)
- HumanoidMimicGen: Data Generation for Loco-Manipulation via Whole-Body Planning· arXiv — cs.RO (Robotics)
- Beyond Binary: Sim-to-Real Dexterous Manipulation with Physics-Grounded Contact Representation· arXiv — cs.RO (Robotics)
- Emerging Extrinsic Dexterity in Cluttered Scenes via Dynamics-aware Policy Learning· arXiv — cs.RO (Robotics)
- BORA: Bridging Offline Reinforcement Learning and Online Residual Adaptation for Real-World Dexterous VLA Models· arXiv — cs.RO (Robotics)
関連記事
More in AI Models
The company just raised its outlook by a staggering amount, and honestly, I'm trying to figure out if this is real momentum or a peak we're about to fall off.
Sarah Williams · 2 hours ago · 5 min
A $65 billion raise that eclipses OpenAI. I've seen big valuations before, but this one's got me scratching my head.
Robert "Bob" Macintosh · 2 hours ago · 3 min
The private equity giants are seeking additional investors for what would be one of the largest AI infrastructure financing deals to date.
James Chen · 3 hours ago · 4 min
The company that once prided itself on vertical integration is outsourcing its AI brain to a competitor. That's not a pivot, it's a concession.



