Science – Page 49

AI Designs AI: A System That Discovers Better Algorithms

26.03.2026 by ebaster

Accuracy diminishes as length increases under conditions of length compression.

Researchers have created an automated framework where artificial intelligence agents independently evolve and refine algorithms for improving language model performance.

Folding Forward: Origami Robots Driven by Mechanical Waves

26.03.2026 by ebaster

A new control system leverages the principles of quadrature oscillation to enable coordinated, electronics-free locomotion in origami-inspired robots.

Seeing is Reacting: Event-Driven Control for Agile Robots

26.03.2026 by ebaster

The system demonstrates closed-loop control of a ground vehicle, utilizing event-based vision to actively stabilize a camera-driven by net event counts from quadratic and linear intensity patterns-at a designated target within a three-dimensional coordinate system defined relative to the robot, camera, and world.

A new approach to robot navigation uses biologically inspired vision to allow robots to react instantly to changes in their environment.

Staging Connection: Robots and the Art of Immersive Drama

26.03.2026 by ebaster

Through interactive role-play scenarios involving three Furhat robots, children engage in a social learning experience designed to foster communication and collaborative skills.

A new look at Robot-Mediated Applied Drama reveals that compelling experiences hinge on crafting a cohesive blend of robotic performance, environmental design, and skillful facilitation.

The Self-Directed Scientist: AI That Designs and Runs Its Own Experiments

26.03.2026 by ebaster

$The system architecture positions a persistent world model [latex]\mathcal{W}[/latex] as a central hub for agent teams, facilitating both collaborative knowledge building and self-correction through a consensus mechanism-detailed in Eq.4-and a review-triggered feedback loop governed by Eq.7, thereby embodying a dynamic and adaptive process of refinement rather than static execution.$

Researchers have developed a new AI framework capable of autonomously formulating research questions, designing experiments, and interpreting results – potentially ushering in an era of curiosity-driven AI discovery.

AI Agents Take the Network to the Testbed

26.03.2026 by ebaster

AgenticNet emerged from an experiment in artificial intelligence, designed not by human programmers, but by an AI coding agent itself-a system built to reverse-engineer its own creation.

A new approach combines the speed of simulation with the realism of emulation, powered by AI coding agents, to unlock more flexible and performant network experimentation.

Outsmarting the Bots: The Power of Human-AI Teams

26.03.2026 by ebaster

The study demonstrates a nuanced alignment between human discernment and artificial intelligence in identifying automated accounts, as evidenced by pairwise agreement rates and Cohen’s κ values-with BotBuster (trained on Twibot-20), RFS (trained on Caverlee-2011), and large language models like llama:70b and mistral-24b exhibiting performance comparable to human judgment in this task.

New research highlights how combining human insight with artificial intelligence can dramatically improve the detection of sophisticated social bots designed to manipulate online conversations.

Beyond Working Together: Charting the Landscape of Robot Collaboration

26.03.2026 by ebaster

As multi-robot systems become increasingly complex, a clear understanding of collaborative behaviors-distinct from simple cooperation or coordination-is crucial for successful deployment.

The Expanding Reach of AI: A Tool-Use Explosion

26.03.2026 by ebaster

The geographic distribution of downloads for action-capable Python packages-analyzed across 6.73 million instances from 528 servers-reveals concentrated activity in North America and Western Europe, with a notable shift in share observed during the latter half of 2025, suggesting evolving patterns of dependency and localized growth within the broader AI ecosystem.

A new analysis of over 177,000 AI agent tools reveals a rapidly growing ‘action space’ and raises critical questions about the future of autonomous systems.

Can AI Really Be a Friend?

26.03.2026 by ebaster

The dataset serves as a foundational record, acknowledging that all systems, even those meticulously constructed, are subject to inevitable degradation and change over time - a principle inherent in their very existence as entities within a temporal medium.

As people increasingly turn to conversational AI for emotional support, researchers are grappling with the ethical implications of forging bonds with machines.