Researchers develop ‘hierarchical AI agent’ that tackles complex errands with ease

ETRI develops
Hierarchical AI Agent. Credit: Electronics and Telecommunications Research Institute (ETRI)

Korean researchers have developed a hierarchical AI technology that autonomously plans even complex, long-horizon tasks. The development of this hierarchical task-planning AI technology, which reduces hallucinations and doubles the success rate, is expected to help robots and agents carry out long-term missions.

The Electronics and Telecommunications Research Institute (ETRI) developed the hierarchical task-planning artificial intelligence (AI) technology “ReAcTree,” which autonomously divides tasks requiring complex and lengthy procedures into subgoals and carries them out, and presented it at AAMAS 2026, one of the world’s premier conferences in the AI agent field.

This research achievem...

Read More

Capturing the cosmic ‘drift’ before a star is born

Capturing the cosmic 'drift' before a star is born
The blue lines represent the magnetic field lines, which are bent due to the gravitational contraction of the core. The red and green dots depict the ion and neutral molecular species, respectively, and the arrows trace their inflow motion towards the core center (the faster they travel the longer the arrows). Credit: Yurika Nakamura and Doris Arzoumanian/Kyushu University

Stars like our sun are formed from the collapse of stellar objects called prestellar cores, cold and dense concentrations of gas and dust held together by gravity. While many questions remain about the exact mechanisms of star formation, advanced radio telescopes have given researchers new insights into the inner workings of infant stars.

Now, publishing in Astronomy & Astrophysics, researchers from Kyushu Univers...

Read More

Meet Biomni—an AI-powered biomedical co-scientist

biomedical research
Credit: Unsplash/CC0 Public Domain

In creating a comprehensive, AI-enabled research agent for the biomedical sciences, Stanford University researchers hope to speed innovation by eliminating the tedium of scientific legwork. Biomni, an AI-powered, multiskilled biomedical research agent, is no mere chatbot. It is a full-fledged “co-scientist” capable of designing and developing complex research workflows, said Jure Leskovec, the Alfred and Rebecca Lin Professor and professor of computer science in the School of Engineering and senior author of the paper introducing Biomni in the journal Science.

“If you think of an agent as a carpenter, a carpenter without tools is just a carpenter who can talk,” Leskovec said, explaining what sets Biomni apart from popular generative AI chatbots...

Read More

These ancient quasars shouldn’t exist so soon after the Big Bang

An artist's concept of a distant galaxy with an active quasar at its center shooting out bright jets of radiation above and below the galactic disc.
Photo Credit
NASA, ESA, Joseph Olmsted (STScI)

Scientists have found the oldest quasars ever seen, revealing giant black holes blazing across the universe when it was only 670 million years old. Astronomers have uncovered 31 of the oldest known quasars, including the two earliest ever detected, shining from a time when the universe was only about 670 million years old. Powered by supermassive black holes billions of times the Sun’s mass, these incredibly bright objects challenge scientists’ understanding of how such enormous black holes formed so quickly after the Big Bang.

Quasars rank among the brightest and most powerful objects in the universe...

Read More