Researchers propose a solution for smart manufacturing using LLM-based agents to generate production sequences, handle runtime faults, and coordinate over MQTT with real-time updates of the factory state.

Researchers introduce ReLiveGym, an evaluation environment for long-lived AI agents that operate over weeks of simulated real-world data. The study investigates how model choice, harness design, and continuous learning affect agent performance on time-sensitive tasks.

This paper proposes a method to improve spatial understanding in LLM-driven agents by combining geometrical tools with LLMs. The approach involves vector-quantizing geodesic trajectories and associating natural language descriptions with them. This allows the LLM to choose the most appropriate tool for a given state and goal, effectively separating learning into two levels: tool discovery and reasoning.

AIMS is a novel AI framework for sim-to-real multi-modal ISAC. It uses a two-agent architecture to generate deployment-specific configurations and coordinates scene construction with task learning. This improves transferability and reduces mismatches among coupled components.

Researchers introduce Interactive-Policy Distillation (IPD), a method that applies adaptive teacher intervention to the student rollout in on-policy distillation, improving performance and data efficiency. IPD trains a student model on its self-generated trajectories with dense token-level teacher feedback, and demonstrates higher accuracy and efficiency compared to traditional on-policy distillation (OPD).

Jagarin is a three-layer architecture for personal AI agents on mobile devices that resolves the deployment paradox through structured hibernation and demand-driven wake. It consists of DAWN, an on-device scoring engine, ARIA, a commercial email identity proxy, and ACE, a protocol for machine-readable communication from institutions to agents.