Cataloguing the boom in agents that reason before they act

What it does This repository is a curated bibliography of research on agentic reasoning—systems where LLMs plan, use tools, search, and learn by interacting with their surroundings. It sorts papers into thematic buckets like foundational planning, self-evolving feedback loops, and collective multi-agent coordination, alongside real-world applications and benchmarks. The list is built to accompany the survey paper Agentic Reasoning for Large Language Models and includes a slide deck.

The interesting bit Rather than dumping every paper into a single chronological firehose, the maintainers impose a three-layer taxonomy: foundational single-agent skills, self-evolving adaptation, and collective multi-agent intelligence. They further tag work by its optimization setting—whether it relies on in-context orchestration or post-training RL and fine-tuning—turning a bibliography into a field guide for an increasingly crowded landscape.

Key highlights

• Mirrors the structure of the arXiv survey (2601.12538) with sections on planning, tool use, search, memory, feedback, and multi-agent systems.
• Covers applied domains including embodied agents, scientific discovery, healthcare, web navigation, and math exploration and vibe coding agents.
• Maintains separate benchmark tables for both core reasoning mechanisms and downstream applications.
• Includes a slide deck summarizing the survey and accepts community contributions via pull requests.

Verdict Worth bookmarking if you are building LLM agents and need to trace the lineage of a specific technique—from tree-search planning to multi-agent role specialization. Skip it if you are looking for executable frameworks or annotated paper summaries; this is strictly an index.