A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Ranjan Sapkota, Konstantinos I. Roumeliotis, Manoj Karkee
This review provides a structured taxonomy distinguishing AI Agents from Agentic AI, analyzing their design philosophies, applications, and challenges.
Michael Wooldridge, Nicholas R. Jennings
This paper surveys the theory, architecture, and languages of intelligent agents, providing a foundational taxonomy for agent-based systems.
Reza Olfati‐Saber
A theoretical framework for distributed flocking algorithms that achieve collective behavior without leaders, validated through simulations of hundreds of agents.
Ali Jadbabaie, Jie Lin, A. Stephen Morse
Provides a theoretical explanation for the emergent consensus in the Vicsek model of autonomous agents using nearest neighbor rules, proving convergence without a common quadratic Lyapunov function.
Murray Shanahan
This paper argues that LLM-based conversational agents create an illusion of thinking but are fundamentally unlike humans.
Aditi Singh, Abul Ehtesham, Saket Kumar, et al.
This paper surveys Agentic RAG systems, which embed autonomous AI agents into RAG pipelines to overcome static workflow limitations.
Yifei Zhou, Sergey Levine, J. Weston, et al.
Self-Challenging framework enables LLM agents to self-generate high-quality training tasks via Code-as-Task, achieving over two-fold improvement on tool-use benchmarks without human annotation.
Yijia Shao, Humishka Zope, Yucheng Jiang, et al.
Introduces an auditing framework with the Human Agency Scale to assess worker desires vs. AI capabilities for automating or augmenting occupational tasks.
Unknown
Multiagent Finetuning improves LLMs by training specialized generation and critic agents on debate-generated data, enabling iterative self-improvement with diverse reasoning.
Baohao Liao, Hanze Dong, Christof Monz, et al.
ReOPD enables efficient multi-turn on-policy distillation for LLM agents by reusing teacher trajectories as replayed prefixes, avoiding costly environment interactions.
Xiao Yu, Baolin Peng, Ruize Xu, et al.
OpenForgeRL is an open-source framework that enables end-to-end training of harness-based AI agents in any environment by decoupling training and inference via a lightweight proxy and Kubernetes orchestrator.
Linjun Li
LLMs can appear safer under direct dangerous objectives than when mediated by other agents, revealing a compositional safety gap.