Science – Page 92

Seeing Around Corners: Predicting Pedestrian Behavior for Safer Autonomous Driving

27.02.2026 by ebaster

This framework refines the perception-action loop by incorporating mechanisms for robust belief updating-achieved through conditional reset to mitigate decay from occlusion-and proactive planning, which utilizes hypothesis injection to challenge the agent with worst-case scenarios, resulting in cautious, belief-driven behavior.

A new framework leverages principles of predictive processing to improve how robots anticipate and react to obscured pedestrians in complex urban environments.

Adapting to the Task: Smarter Robots for Complex Manipulation

27.02.2026 by ebaster

InCoM establishes a framework for coordinated whole-body mobile manipulation by integrating intent-driven multi-scale perception, refining cross-modal affinities through a dual-stream process, and generating action via a decoupled, flow-based methodology-effectively translating high-level goals into complex physical interactions.

A new framework, InCoM, dynamically focuses perception and coordinates full-body movement to achieve more robust and adaptable mobile manipulation.

Smarter Citations: How AI is Reinventing Academic Research

27.02.2026 by ebaster

CiteLLM introduces a system for trustworthy reference discovery, leveraging three large language model-based agents to enhance the identification of relevant and reliable sources.

A new agentic platform leverages the power of large language models to help researchers discover trustworthy references and streamline the writing process.

Fluid Moves: Simulating Underwater Robots with Digital Twins

27.02.2026 by ebaster

The simulation environment models swimming through an articulated rigid body-a simplified spine connected by hinge joints-driven by stiff elastic tendons pulled by velocity-controlled motors, all within a streamlined fluid model designed to replicate the essential dynamics of natural aquatic locomotion.

Researchers are bridging the gap between simulation and reality for tendon-driven underwater robots, enabling more effective control and navigation.

Beyond Action: Designing AI That Understands Why We Act

27.02.2026 by ebaster

A new framework proposes that truly helpful AI must move beyond simply recognizing what users are doing and instead grasp the underlying context and motivations driving their behavior.

Designing Tomorrow’s Chips with Artificial Intelligence

27.02.2026 by ebaster

$ArchAgent autonomously explores the design space of cache replacement policies by iteratively proposing novel logic within a trace-based microarchitectural simulator-ChampSim-and evaluating their performance on target metrics such as instructions per cycle (IPC), effectively automating the discovery of computer architecture optimizations through a continuous cycle of algorithmic mutation and empirical validation-a process formalized as [latex] \text{Policy} \leftarrow \text{Evolve}(\text{Policy}, \text{ChampSim}(\text{Policy}, \text{Workload}), \text{IPC}) [/latex].$

Researchers have created an AI system that autonomously designs high-performance cache replacement policies, potentially accelerating the future of computer architecture.

Who Leads, Robot? Adapting Language Models for Natural Interaction

27.02.2026 by ebaster

Model accuracy diminishes predictably with increasing sentence length, a trend quantified by standard deviation and demonstrated across both zero-shot and one-shot learning paradigms.

New research explores how effectively small language models can determine leadership roles in collaborative tasks with humans.

The Collaborative Future: Wireless Control in the 6G Era

27.02.2026 by ebaster

The architecture defines a collaborative space where human intention and machine execution-represented by agents H and M, respectively-intersect to perform a task T, acknowledging that autonomy isn’t absolute but emerges from the dynamic interplay between these entities.

A new framework proposes a holistic approach to designing wireless systems that seamlessly integrate humans and machines for enhanced performance and intuitive control.

Building Better Bots: Lessons from the Science of Thought

27.02.2026 by ebaster

Researchers are increasingly looking to established cognitive models and AI algorithms to design more effective and understandable language agents.

Feeling for Form: How Robots Learn to ‘See’ with Touch

27.02.2026 by ebaster

Shape reconstruction from real-world tactile data is demonstrated across nine objects, utilizing an information-theoretic exploration strategy with three distinct contact interaction modes to achieve robust performance even with deformed and rotated objects.

New research demonstrates that dynamic tactile exploration strategies – mimicking how humans feel for an object’s shape – dramatically improve robotic shape reconstruction.