Skip to content

AI Dynamics

Global AI News Aggregator

Rechercher

@mit_csail

PRoC3S Successfully Demonstrates Drawing and Block Stacking Tasks

By

@mit_csail

–

12 December 2024 18h08

In simulations, PRoC3S successfully drew stars & letters 8/10 times each. It also could stack digital blocks in pyramids & lines, & place items w/accuracy, like fruits on a plate.

Across each of these digital demos, the CSAIL method completed the requested task more… pic.twitter.com/lW9Q1HC0Mp
— MIT CSAIL (@MIT_CSAIL) 12 décembre 2024

In simulations, PRoC3S successfully drew stars & letters 8/10 times each. It also could stack digital blocks in pyramids & lines, & place items w/accuracy, like fruits on a plate. Across each of these digital demos, the CSAIL method completed the requested task more

→ View original post on X — @mit_csail

12 December 2024
Teaching Robots Safe Task Execution Using Vision and LLMs

By

@mit_csail

–

12 December 2024 18h08

Can we teach a robot its limits to do chores safely & correctly? 🧵

To help robots execute open-ended, multi-step tasks, MIT CSAIL researchers used vision models to see what’s near the machine & model its constraints. An LLM sketches up a plan that’s checked in a simulator to… pic.twitter.com/u5HyfQwUic
— MIT CSAIL (@MIT_CSAIL) 12 décembre 2024

Can we teach a robot its limits to do chores safely & correctly? To help robots execute open-ended, multi-step tasks, MIT CSAIL researchers used vision models to see what’s near the machine & model its constraints. An LLM sketches up a plan that’s checked in a simulator to

→ View original post on X — @mit_csail

12 December 2024
PRoC3S: Robot Planning Via Code for Constraint Satisfaction

By

@mit_csail

–

12 December 2024 18h08

Their trial-&-error method is called "Planning for Robots via Code for Continuous Constraint Satisfaction" (PRoC3S). It tests long-horizon plans to ensure they satisfy all constraints & enables a robot to perform diverse tasks, like writing individual letters, drawing a star, &

→ View original post on X — @mit_csail

12 December 2024
AI Models Confidently Make Mistakes Despite External Context

By

@mit_csail

–

11 December 2024 18h39

AI systems often gather external info as context when answering a particular query. Even w/this relevant info, models can make mistakes w/seemingly high doses of confidence.

→ View original post on X — @mit_csail

11 December 2024
ContextCite: Tracing AI Sources and Detecting Hallucinations

By

@mit_csail

–

11 December 2024 18h39

If the AI generates an inaccurate fact, users can trace the error back to its original source & understand the model’s reasoning. If the AI hallucinates an answer, ContextCite can indicate that the information didn’t come from any real source at all.

The tool could be valuable… pic.twitter.com/D8ViuL2cjX
— MIT CSAIL (@MIT_CSAIL) 11 décembre 2024

If the AI generates an inaccurate fact, users can trace the error back to its original source & understand the model’s reasoning. If the AI hallucinates an answer, ContextCite can indicate that the information didn’t come from any real source at all. The tool could be valuable

→ View original post on X — @mit_csail

11 December 2024
ContextCite helps verify AI assistant claims with source citations

By

@mit_csail

–

11 December 2024 18h39

Existing AI assistants sometimes provide source links, but users would have to tediously review the article to spot any mistakes. ContextCite can help directly find the specific sentence that a model used, making it easier to verify claims & spot mistakes.

→ View original post on X — @mit_csail

11 December 2024
MIT ContextCite tool improves chatbot answer reliability verification

By

@mit_csail

–

11 December 2024 18h39

How can we really know if a chatbot is giving a reliable answer? MIT CSAIL’s "ContextCite" tool can ID the parts of external context used to generate any particular statement from a language model, improving trust by helping users easily verify the statement:

→ View original post on X — @mit_csail

11 December 2024
Ada Lovelace: Pioneer of Computing and Programming History

By

@mit_csail

–

10 December 2024 18h00

Born #otd in 1815: Ada Lovelace, the woman often credited with creating "the world's first program": https://
bit.ly/3ZvJhIx

→ View original post on X — @mit_csail

10 December 2024
MIT Professor Patrick Winston’s Simple Neural Network Breakdown

By

@mit_csail

–

09 December 2024 18h00

A breakdown of the simplest neural network from the late MIT Prof. Patrick Winston during his 6.034 (Artificial Intelligence) course.

Clip v/@7etsuo pic.twitter.com/BoUl7iDbbn
— MIT CSAIL (@MIT_CSAIL) 9 décembre 2024

A breakdown of the simplest neural network from the late MIT Prof. Patrick Winston during his 6.034 (Artificial Intelligence) course. Clip v/
@7etsuo

→ View original post on X — @mit_csail

9 December 2024
Breaking Down Problems: Essential Skill in Computer Science

By

@mit_csail

–

08 December 2024 18h00

"Learning to break down problems into smaller pieces is one of the most important skills in computer science/life." — Addy Osmani (
@addyosmani
)

→ View original post on X — @mit_csail

8 December 2024

←Previous Page

1 … 35 36 37 38 39 … 82

INNOVATION GENERATIVE AI RESEARCH LLMS TOOLS MACHINE LEARNING CODE MARKET TRENDS TECHNOLOGY BUSINESS BIG TECH ETHICS ENTERPRISE AI SOFTWARE AGENTS AUTOMATION APPS COMPUTING DATA POLICY OPEN SOURCE MULTIMODAL AI REGULATION CULTURE CREATIVE AI PROMPT ENGINEERING SOCIETY ECONOMY SAFETY EDUCATION INVESTMENT AI HARDWARE AGI HARDWARE JOBS STARTUPS INDUSTRY ROBOTICS WORKFORCE SECURITY CYBERSECURITY HEALTHCARE AI SYSTEMS SUSTAINABILITY WEB3 DECENTRALIZED AI

AI Dynamics

Global AI News Aggregator

About
Archives
Contact

Rechercher