Archives des Computer science and technology

Ecologists find computer vision models’ blind spots in retrieving wildlife images
22 Déc
Need a research hypothesis? Ask AI.
22 Déc
MIT researchers introduce Boltz-1, a fully open-source model for predicting biomolecular structures
22 Déc
Study reveals AI chatbots can detect race, but racial bias reduces response empathy
22 Déc
Teaching a robot its limits, to complete open-ended tasks safely
22 Déc
AI in health should be regulated, but don’t forget about the algorithms, researchers say
22 Déc
Researchers reduce bias in AI models while preserving or improving accuracy
22 Déc
Study: Some language reward models exhibit political bias
22 Déc
Enabling AI to explain its predictions in plain language
22 Déc
Daniela Rus wins John Scott Award
22 Déc
Citation tool offers a new approach to trustworthy AI-generated content
22 Déc
A new way to create realistic 3D shapes using generative AI
22 Déc
Photonic processor could enable ultrafast AI computations with extreme energy efficiency
22 Déc
Improving health, one machine learning system at a time
22 Déc
MIT researchers develop an efficient way to train more reliable AI agents
22 Déc
Advancing urban tree monitoring with AI-powered digital twins
22 Déc
Can robots learn from machine dreams?
22 Déc
Four from MIT named 2025 Rhodes Scholars
22 Déc
Graph-based AI model maps the future of innovation
22 Déc
3 Questions: Inverting the problem of design
22 Déc
A causal theory for studying the cause-and-effect relationships of genes
22 Déc
A portable light system that can digitize everyday objects
22 Déc
Despite its impressive output, generative AI doesn’t have a coherent understanding of the world
22 Déc
Nanoscale transistors could enable more efficient electronics
22 Déc
A faster, better way to train general-purpose robots
22 Déc
Making it easier to verify an AI model’s responses
22 Déc
Combining next-token prediction and video diffusion in computer vision and robotics
22 Déc
Modeling relationships to solve complex problems efficiently
22 Déc
How AI is improving simulations with smarter sampling techniques
22 Déc
AI simulation gives people a glimpse of their potential future self
22 Déc
AI pareidolia: Can machines spot faces in inanimate objects?
22 Déc
MIT launches new Music Technology and Computation Graduate Program
22 Déc
3 Questions: Should we label AI systems like we do prescription drugs?
22 Déc
Study: AI could lead to inconsistent outcomes in home surveillance
22 Déc
Enhancing LLM collaboration for smarter, more efficient solutions
22 Déc
A fast and flexible approach to help doctors annotate medical scans
22 Déc
Study: Transparency is often lacking in datasets used to train large language models
22 Déc
A framework for solving parabolic partial differential equations
22 Déc
LLMs develop their own understanding of reality as their language abilities improve
22 Déc
MIT researchers use large language models to flag problems in complex systems
22 Déc
Helping robots practice skills independently to adapt to unfamiliar environments
22 Déc
Helping Olympic athletes optimize their performance, one stride at a time
22 Déc
Study: When allocating scarce resources with AI, randomization can improve fairness
22 Déc
MIT researchers advance automated interpretability in AI models
22 Déc
Large language models don’t behave like people, even though we may expect them to
22 Déc
AI model identifies certain breast tumor stages likely to progress to invasive cancer
22 Déc
Creating and verifying stable AI-controlled systems in a rigorous and flexible way
22 Déc
AI method radically speeds predictions of materials’ thermal properties
22 Déc
How to assess a general-purpose AI model’s reliability before it’s deployed
22 Déc
Marking a milestone: Dedication ceremony celebrates the new MIT Schwarzman College of Computing building
22 Déc
Reasoning skills of large language models are often overestimated
22 Déc
When to trust an AI model
22 Déc
“They can see themselves shaping the world they live in”
22 Déc
MIT researchers introduce generative AI for databases
22 Déc
Researchers leverage shadows to model 3D scenes, including objects blocked from view
22 Déc
Understanding the visual knowledge of language models
22 Déc
A smarter way to streamline drug discovery
22 Déc
Technique improves the reasoning capabilities of large language models
22 Déc
Researchers use large language models to help robots navigate
22 Déc
New algorithm discovers language just by watching videos
22 Déc
Mouth-based touchpad enables people living with paralysis to interact with computers
22 Déc
A technique for more effective multipurpose robots
22 Déc
Looking for a specific action in a video? This AI-based method can find it for you
22 Déc
Controlled diffusion model can change material properties in images
22 Déc
School of Engineering welcomes new faculty
22 Déc
Scientists use generative AI to answer complex questions in physics
22 Déc
Using ideas from game theory to improve the reliability of language models
22 Déc
A better way to control shape-shifting soft robots
22 Déc
Creating bespoke programming languages for efficient visual AI systems
22 Déc
Natural language boosts LLM performance in coding, planning, and robotics
22 Déc
Mapping the brain pathways of visual memorability
22 Déc
This tiny chip can safeguard user data while enabling efficient computing on a smartphone
22 Déc
To build a better AI helper, start by modeling the irrational behavior of humans
22 Déc

Ecologists find computer vision models’ blind spots in retrieving wildlife images

Need a research hypothesis? Ask AI.

MIT researchers introduce Boltz-1, a fully open-source model for predicting biomolecular structures

Study reveals AI chatbots can detect race, but racial bias reduces response empathy

Teaching a robot its limits, to complete open-ended tasks safely

AI in health should be regulated, but don’t forget about the algorithms, researchers say

Researchers reduce bias in AI models while preserving or improving accuracy

Study: Some language reward models exhibit political bias

Enabling AI to explain its predictions in plain language

Daniela Rus wins John Scott Award

Citation tool offers a new approach to trustworthy AI-generated content

A new way to create realistic 3D shapes using generative AI

Photonic processor could enable ultrafast AI computations with extreme energy efficiency

Improving health, one machine learning system at a time

MIT researchers develop an efficient way to train more reliable AI agents

Advancing urban tree monitoring with AI-powered digital twins

Can robots learn from machine dreams?

Four from MIT named 2025 Rhodes Scholars

Graph-based AI model maps the future of innovation

3 Questions: Inverting the problem of design

A causal theory for studying the cause-and-effect relationships of genes

A portable light system that can digitize everyday objects

Despite its impressive output, generative AI doesn’t have a coherent understanding of the world

Nanoscale transistors could enable more efficient electronics

A faster, better way to train general-purpose robots

Making it easier to verify an AI model’s responses

Combining next-token prediction and video diffusion in computer vision and robotics

Modeling relationships to solve complex problems efficiently

How AI is improving simulations with smarter sampling techniques

AI simulation gives people a glimpse of their potential future self

AI pareidolia: Can machines spot faces in inanimate objects?

MIT launches new Music Technology and Computation Graduate Program

3 Questions: Should we label AI systems like we do prescription drugs?

Study: AI could lead to inconsistent outcomes in home surveillance

Enhancing LLM collaboration for smarter, more efficient solutions

A fast and flexible approach to help doctors annotate medical scans

Study: Transparency is often lacking in datasets used to train large language models

A framework for solving parabolic partial differential equations

LLMs develop their own understanding of reality as their language abilities improve

MIT researchers use large language models to flag problems in complex systems

Helping robots practice skills independently to adapt to unfamiliar environments

Helping Olympic athletes optimize their performance, one stride at a time

Study: When allocating scarce resources with AI, randomization can improve fairness

MIT researchers advance automated interpretability in AI models

Large language models don’t behave like people, even though we may expect them to

AI model identifies certain breast tumor stages likely to progress to invasive cancer

Creating and verifying stable AI-controlled systems in a rigorous and flexible way

AI method radically speeds predictions of materials’ thermal properties

How to assess a general-purpose AI model’s reliability before it’s deployed

Marking a milestone: Dedication ceremony celebrates the new MIT Schwarzman College of Computing building

Reasoning skills of large language models are often overestimated

When to trust an AI model

“They can see themselves shaping the world they live in”

MIT researchers introduce generative AI for databases

Researchers leverage shadows to model 3D scenes, including objects blocked from view

Understanding the visual knowledge of language models

A smarter way to streamline drug discovery

Technique improves the reasoning capabilities of large language models

Researchers use large language models to help robots navigate

New algorithm discovers language just by watching videos

Mouth-based touchpad enables people living with paralysis to interact with computers

A technique for more effective multipurpose robots

Looking for a specific action in a video? This AI-based method can find it for you

Controlled diffusion model can change material properties in images

School of Engineering welcomes new faculty

Scientists use generative AI to answer complex questions in physics

Using ideas from game theory to improve the reliability of language models

A better way to control shape-shifting soft robots

Creating bespoke programming languages for efficient visual AI systems

Natural language boosts LLM performance in coding, planning, and robotics

Mapping the brain pathways of visual memorability

This tiny chip can safeguard user data while enabling efficient computing on a smartphone

To build a better AI helper, start by modeling the irrational behavior of humans