Jul 28, 2026
Efficiency Matters in Autonomous Research
Efficiency Matters in Autonomous Research AI-driven autonomous research (AR) systems are increasingly used to solve complex problems in science and engineeri...
Jul 27, 2026
Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture
Dynamic Capability Scoping for Enterprise AI Agents: A Synthetic Dataset and Three-Source Permission Architecture Enterprise AI agents are often given a "sta...
Jul 27, 2026
IDEAgent: Agentic Quality-Diversity Search for Research Idea Generation
IDEAgent: Agentic Quality-Diversity Search for Research Idea Generation Large Language Models (LLMs) are increasingly used to assist in scientific discovery,...
Jul 27, 2026
TRACE-ROUTER: Task-Consistent and Adaptive Online Routing for Agentic AI
TRACE-Router: Task-Consistent and Adaptive Online Routing for Agentic AI Modern enterprise AI often relies on a mix of large, powerful models and smaller, fa...
Jul 27, 2026
Explainable Reinforcement Learning for assisting Air Traffic Controllers
Explainable Reinforcement Learning for assisting Air Traffic Controllers This research explores how to make Reinforcement Learning (RL) systems more transpar...
Jul 27, 2026
The Regression Tax: Decomposing Why Skills Help and Hurt LLM Agents
The Regression Tax: Decomposing Why Skills Help — and Hurt — LLM Agents This paper investigates a hidden cost in AI development: the tendency for "procedural...
Jul 27, 2026
Do Agent Benchmarks Measure Capability? Protocol Validity in the Age of Agentic AI
This paper investigates whether current AI agent benchmarks actually measure the capabilities they claim to test.
Jul 24, 2026
Regulating autonomous and agentic AI
Regulating autonomous and agentic AI The paper "Regulating autonomous and agentic AI" examines the growing disconnect between traditional regulatory framewor...
Jul 24, 2026
OpenForgeRL: Train Harness-native Agents in Any Environment
Modern AI agents are increasingly powerful, but they are often wrapped in complex "inference harnesses"—software scaffolds that manage multi-turn reasoning,...
Jul 24, 2026
ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders
ICAE-Bench: Evaluating Coding Agents as Interactive Project Builders The rise of "vibe-coding"—where developers start with a high-level idea rather than a de...