ai agent papers

Report Abuse

Basic Information

AI Agent Papers is a curated, continuously updated bibliography and reading guide that collects recent academic and technical papers about autonomous AI agents. The repository aggregates monthly highlights and themed compilations covering topics such as agent blueprints, applications, enterprise agents, data agents, research agents, role playing agents, memory systems, multi-agent systems, inference-time computing, tool-integrated reasoning, self-improvement and reinforcement learning for agents. It organizes entries by month with short highlights and links to individual papers hosted mainly on arXiv and other preprint servers. The README also enumerates paper categories and provides structured subpages for capabilities, architectures, applications, and presentations to help readers navigate the literature. The collection is maintained biweekly and lists recommended, survey, and benchmark papers where indicated.

Links

App Details

Features
Regular biweekly updates and monthly highlights that surface recent and notable agent research. The README groups papers into clear topical sections such as Agent Blueprints, Agent Applications, Enterprise Agents, Data Agents, Memory, Multi-Agent, Deep Research Agents, and more. Dedicated paper category pages cover agent capabilities like planning, reasoning, profile, perception, tool use, self-correction, memory, safety, tuning, and evaluation. Architecture and application indexes separate single-agent and multi-agent frameworks and list embodied, digital, web, mobile, and software agent applications. The repo annotates items with recommendation, survey, and benchmark markers and includes a workflow image and references to related curated lists. It links to individual papers and organizes material for quick discovery and reading.
Use Cases
The repository helps researchers, practitioners, and students locate and track recent advances in AI agent research without searching multiple sources. It provides thematic reading lists and monthly highlights to prioritize important works, plus structured category pages for deep dives into specific agent capabilities and architectures. The curated layout supports literature reviews, syllabus design, and staying current on topics such as tool integration, memory systems, reinforcement learning for agents, multi-agent collaboration, and enterprise use cases. Annotated markers for surveys and benchmarks assist users in finding comprehensive overviews and evaluation resources. Cross-references and a short references section point to other collections, making this a compact starting point for exploring academic and applied developments in agentic AI.

Please fill the required fields*