Adam Zewe

Machine-learning framework speeds predictions for material thermal properties

MIT researchers have developed a machine-learning framework designed to help engineers design more efficient energy-conversion systems and faster microelectronic devices, reducing waste heat.

Using large language models to help robots navigate

Researchers developed a method that uses language-based inputs instead of costly visual data to direct a robot through a multistep navigation task.

Controlling shape-shifting soft robots effectively

An algorithm learns to squish, bend, or stretch a robot’s entire body to accomplish diverse tasks like avoiding obstacles or retrieving items.

Qubit array tuned for better quantum computing results

The advance developed by MIT scientists offers a way to characterize a fundamental resource needed for quantum computing.

Computer chip safeguards smart phone user data while enabling efficiency

Researchers have developed a security solution for power-hungry AI models that offers protection against two common attacks.

Algorithm helps robots navigate efficiently in uncertain environments

A new algorithm reduces travel time by identifying shortcuts a robot could take on the way to its destination.

Peripheral vision in AI models has been enhanced

By enabling AI models to see the world more like humans do, the work could help improve driver safety and shed light on human behavior.

2D magnetic materials harnessed for energy-efficient computing

MIT researchers precisely controlled an ultrathin magnet at room temperature, which could enable faster, more efficient computers.

Tamper-proof ID tag can authenticate almost anything

MIT engineers developed a tag that can reveal with near-perfect accuracy whether an item is real or fake.

Technique could enhance quantum sensing device sensitivity

MIT researchers have developed a method that helps identify and control larger numbers of atomic-scale defects, to build a bigger system of qubits.

How generative AI works and helps engineers

How do powerful generative artificial intelligence (AI) systems like ChatGPT work, and what makes them different from other types of AI?

How language gives robots a better grasp of an open-ended world

By blending 2D images with foundation models to build 3D feature fields, a new MIT method helps robots understand and manipulate nearby objects with open-ended language prompts.

Techniques efficiently accelerate sparse tensors for generative AI models

Researchers from MIT and NVIDIA developed two complementary techniques that could dramatically boost the speed and performance of computing applications such as generative AI.

Generative AI technique helps robots pack objects into a tight space

MIT researchers developed a technique that coaxed a family of generative AI models to work together to solve multistep robot manipulation problems.

Finger-shaped sensor enables more dexterous robots

MIT engineers develop a long, curved touch sensor that could enable a robot to grasp and manipulate objects in multiple ways.

Qubit circuit enables quantum operations with higher accuracy

MIT researchers' advances help bring quantum error correction a step closer to reality and make them almost 100% accurate.

AI technique helps robots manipulate objects with their whole bodies

A technique has been developed that lets a robot to reason efficiently about moving objects using more than just its fingertips. See video.

Machine-learning method developed to make controlling robots easier

Researchers developed a machine-learning technique that can efficiently learn to control a robot, leading to better performance with less data required.

Teaching robots to perform tasks faster

MIT researchers developed a framework that helps a nontechnical user understand why a robot failed and fine-tune it with minimal effort to perform a task effectively.

Precise nanoLED arrays are grown, developed

MIT researchers have developed nanoLED arrays that could be beneficial for quantum light sources, and augmented and virtual reality (AR/VR) displays.

AI-enabled approach helps make autonomous robots safe, reliable

Researchers have developed an AI-based approach for controlling autonomous robots that satisfies the often-conflicting goals of safety and stability.

Using reflections to enhance perspectives and vision

A computer vision system has been developed to turn any shiny object into a camera of sorts, enabling an observer to see around corners or beyond obstructions.

Training machines to learn more like humans do

MIT researchers have identified a property that helps computer vision models learn to represent the visual world in a more stable way.

MIT engineers grow atomically thin transistors on top of computer chips

Low-temperature growth and fabrication technology allows the integration of 2D materials directly onto a silicon circuit, which could lead to denser and more powerful computer chips.

Robotic hand identifies objects after one grasp

MIT researchers have developed a three-fingered robotic gripper that can “feel” with great sensitivity along the full length of each finger.

Data transmission chip for decoding demonstrates strong energy efficiency

The chip developed by MIT researchers, which can decipher any encoded signal, could enable lower-cost devices that perform better while requiring less hardware.

Wireless technique helps quantum computing systems keep their cool

MIT researchers have developed a wireless technique that enables a super-cold quantum computer to send and receive data without generating too much error-causing heat.

Technique developed to improve machine-learning models’ reliability

MIT researchers have developed a machine-learning model to determine its confidence in a prediction while using no additional data.

Boosting quantum computing signals while reducing noise

Squeezing noise over a broad frequency bandwidth in a quantum computing system could lead to faster and more accurate quantum computing measurements.

Neural network models learning from examples without training

An MIT study shows how large language models can learn a new task from just a few examples, without the need for any new training data.

Quantum computing architecture designed to connect large-scale devices

MIT researchers have demonstrated directional photon emission, the first step toward extensible quantum interconnects, which could help connect large-scale devices.

Paper-thin solar cell can turn surfaces into power source

MIT researchers develop a scalable fabrication technique to produce ultrathin solar cells that can be seamlessly added to any surface. See video.

Technique developed to help 2D materials expand

A technique that accurately measures how atom-thin materials expand when heated could help engineers develop faster, more powerful electronic devices.

Machine learning facilitates turbulence tracking in fusion reactors

MIT researchers have developed a machine learning approach that can affect the energy generated during fusion reactions, with implications for reactor design.

Improving deep learning with light photonics

A new method uses optics to accelerate machine-learning computations on smart speakers and other low-power connected devices.

Collaborative machine learning preserves privacy

Federated learning is a collaborative method for training a machine-learning model that keeps sensitive user data private.

Soft actuators in robots emit light in colors, patterns

MIT researchers have created insect-scale robots inspired by fireflies that can emit light when they fly, which enables motion tracking and communication. See video.

Stronger security systems for smart devices

MIT researchers demonstrate two security methods that efficiently protect analog-to-digital converters from powerful attacks that aim to steal user data.

Easier ways to teach robots new skills

MIT researchers have developed a technique that enables a robot to learn a pick-and-place task with only a handful of human demonstrations.

Teaching robots to manipulate soft objects

Researchers have developed a technique that could enable a robot to manipulate squishy objects like pizza dough or soft materials like clothing.

Quantum circuits developed for noise resiliency

Researchers are working on developing a technique that makes the quantum circuit itself resilient to noise to improve quantum computing's performance.

Helping humans understand robots

Researchers from MIT and Harvard suggest that applying theories from cognitive science and educational psychology to the area of human-robot interaction can help humans build more accurate mental models of their robot collaborators.

Peripheral vision benefits for machines

MIT researchers have found similarities between how some computer-vision systems process images and how humans see out of the corners of their eyes.

Tiny materials lead to quantum computing advance

Using ultrathin materials to reduce the size of superconducting qubits may help make personal-sized quantum computing devices according to MIT researchers.

Artificial intelligence developed to understand object relationships

A machine-learning model developed by MIT researchers could enable robots to understand interactions in the world in the way humans do with artificial intelligence.

Giving robots social skills

A machine-learning system developed by MIT researchers can help robots understand and perform certain social interactions.

Avoiding shortcut solutions in artificial intelligence

A method developed by MIT researchers forces a machine-learning model to focus on more data when learning a task, which leads to more reliable predictions.

Control system enables four-legged robot to manage uneven terrain

MIT researchers have developed a system that improves the speed and agility of four-legged robots as they jump across gaps in the terrain. See video.

Using blockchain technology to protect robots

Blockchain technology as a communication tool for a team of robots could provide security and safeguard against deception.

Robot digs, finds objects under piles

An integrated robotic arm fuses visual data from a camera and radio frequency (RF) information from an antenna to find and retrieve objects, even when they are buried under a pile and fully out of view. See video.