Content by Adam Zewe
Robotic Helper Making Mistakes? Just Nudge ItNew research could enable a person to correct a robot’s actions in real time using the kind of feedback they’d give another human
Wed, 03/26/2025 - 12:02
Imagine that a robot is helping you clean the dishes. You ask it to grab a soapy bowl out of the sink, but its gripper slightly misses the mark. Using a new framework developed by MIT and NVIDIA researchers, you could correct that robot’s behavior…
Fast, Agile Robotic Insect Could Aid Mechanical PollinationNewly designed bug-sized bot flies 100 times longer than prior versions
Thu, 02/06/2025 - 00:03
With a more efficient method for artificial pollination, farmers in the future could grow fruits and vegetables inside multilevel warehouses, boosting yields while mitigating some of agriculture’s harmful effects on the environment. To help make…
Despite Impressive Output, Generative AI Lacks Understanding of the WorldEven the best-performing large language models don’t form a true model
Mon, 11/18/2024 - 12:02
Large language models can do impressive things, like write poetry or generate viable computer programs, even though these models are trained to predict words that come next in a piece of text. Such surprising capabilities can make it seem like the…
A Faster, Better Way to Train General-Purpose RobotsResearchers develop a technique that pools diverse data to teach new skills
Thu, 11/07/2024 - 12:02
In the 1960s cartoon series The Jetsons, Rosie the robotic maid seamlessly switches from vacuuming the house to cooking dinner to taking out the trash. But in real life, training a general-purpose robot remains a major challenge. Typically,…
Making It Easier to Verify an AI Model’s Responses This tool helps spot AI errors
Wed, 11/06/2024 - 12:03
Despite their impressive capabilities, large language models (LLMs) are far from perfect. These artificial intelligence models sometimes “hallucinate” by generating incorrect or unsupported information in response to a query. Due to this…
Using Large Language Models to Flag Problems in Complex SystemsApproach can detect anomalies in data recorded over time with no need for training
Wed, 09/04/2024 - 12:02
Identifying one faulty turbine in a wind farm, which can involve looking at hundreds of signals and millions of data points, is akin to finding a needle in a haystack. Engineers often streamline this complex problem using deep-learning models that…
Researchers Leverage Shadows to Model 3D ScenesTechnique could lead to safer autonomous vehicles or faster warehouse robots
Wed, 07/03/2024 - 23:58
(MIT: Cambridge, MA) -- Imagine driving through a tunnel in an autonomous vehicle but, unbeknownst to you, a crash has stopped traffic up ahead. Normally, you’d need to rely on the car in front of you to know you should start braking. But what if…
Researchers Use Large Language Models to Help Robots NavigateWords instead of costly visual data direct multistep navigation tasks
Wed, 07/03/2024 - 12:02
Someday, you may want your home robot to carry a load of dirty clothes downstairs and deposit them in the washing machine in the far-left corner of the basement. The robot will need to combine your instructions with its visual observations to…
Researchers Demonstrate First Chip-Based 3D PrinterSmaller than a coin, optical device could enable rapid prototyping on the go
Wed, 06/19/2024 - 12:01
Imagine a portable 3D printer you could hold in the palm of your hand. This tiny device could enable you to rapidly create customized, low-cost objects on the go, like a fastener to repair a wobbly bicycle wheel or a component for a critical medical…
3D Printer Can Figure Out How to Print With Unknown MaterialUsing renewable or recyclable materials that are more sustainable
Tue, 06/18/2024 - 12:00
(MIT News: Cambridge, MA) – While 3D printing has exploded in popularity, many of the plastic materials these printers use to create objects can’t be easily recycled. New sustainable materials are emerging for use in 3D printing, but they remain…