القائمة
Curated research collection on adaptation strategies for agentic AI systems, covering agent and tool adaptation methods with RL, SFT, and DPO approaches
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Train and deploy Qwen-AgentWorld, a native language world model that simulates agentic environments across 7 domains (MCP, Search, Terminal, SWE, Android, Web, OS) for agent training and evaluation.
Build and orchestrate collaborative multi-agent teams using HiClaw's Manager-Workers architecture on Matrix with Kubernetes-native control.
Set up and manage collaborative multi-agent teams using HiClaw, a Kubernetes-native platform with Matrix rooms for human-in-the-loop AI coordination
Build and orchestrate collaborative multi-agent teams using HiClaw's Manager-Workers architecture with Matrix rooms, MCP servers, and Kubernetes-native deployment.
Deploy and orchestrate collaborative multi-agent teams using HiClaw's Manager-Workers architecture on Docker or Kubernetes with Matrix rooms for human oversight
Use Agent Apprenticeship to train AI agents through real-world tasks, reusable experience, and ecosystem learning signals
| name | awesome-adaptation-agentic-ai |
| description | Curated research collection on adaptation strategies for agentic AI systems, covering agent and tool adaptation methods with RL, SFT, and DPO approaches |
| triggers | ["find papers on agentic AI adaptation","how do AI agents adapt their tool use","research on tool execution adaptation","agent reinforcement learning methods","lookup agentic AI adaptation techniques","show me agent adaptation frameworks","what are tool adaptation strategies","find research on adaptive AI agents"] |
Skill by ara.so — AI Agent Skills collection.
This repository is a curated research collection accompanying the paper "Adaptation of Agentic AI" (arXiv:2512.16301). It systematically organizes papers on how AI agents adapt their behavior, particularly focusing on:
The collection categorizes 40+ research papers by adaptation strategy, training method (RL, SFT, DPO), task domain, and implementation details.
awesome-adaptation-of-agentic-ai/
├── README.md # Main paper collection with categorized tables
├── images/
│ ├── intro.png # Overview diagram
│ ├── a1_illustrate.png # Tool execution signaled adaptation
│ ├── a1_timeline.png # Development timeline
│ ├── paper_icon.png # Paper link icons
│ └── code_icon.png # Code link icons
└── LICENSE
# Clone the repository
git clone https://github.com/pat-jj/Awesome-Adaptation-of-Agentic-AI.git
cd Awesome-Adaptation-of-Agentic-AI
# View the main README
cat README.md
The repository organizes papers into four main categories:
A1: Tool Execution Signaled Agent Adaptation
A2: Agent Output Signaled Adaptation
T1: Agent-Agnostic Tool Adaptation
T2: Agent-Supervised Tool Adaptation
Papers where agents learn from tool execution feedback using reinforcement learning:
Notable Methods:
Common Pattern:
Agent → Tool Call → Execution Feedback → RL Update (GRPO/PPO)
Papers using supervised fine-tuning and direct preference optimization:
Notable Methods:
Common Pattern:
Agent → Tool Call → Execution Trace → Supervised Learning
Example: Formal Theorem Proving
# Search README for theorem proving papers
grep -i "theorem proving" README.md
Papers include: AlphaProof, BFS-Prover-V2, Goedel-Prover-V2, Leanabell-Prover-V2, DeepSeek-Prover-V1.5/V2
Example: Coding & Code Execution
# Find coding-related papers
grep -i "coding\|code executor" README.md
Papers include: olmOCR2, R1-Code-Interpreter, Code-R1, DeepSeek-R1-Zero, RLEF, LeDex, CYCLE, CodeAct
Example: GRPO (Group Relative Policy Optimization)
# List all GRPO papers
grep "GRPO" README.md
Commonly used in: Tool-N1, DeepSeek-Prover-V2, SQL-R1, Rec-R1, DeepRetrieval, etc.
Example: DPO (Direct Preference Optimization)
# List all DPO papers
grep "DPO" README.md
Used in: AWL, LeReT, TP-LLaMA, RetPO
Example: Qwen2.5-based agents
# Find Qwen2.5 implementations
grep "Qwen2.5" README.md
Models include: olmOCR2, ToolExpander, BFS-Prover-V2, WebGen-Agent, Tool-R1, etc.
All papers include arXiv or conference links:
[Paper](https://arxiv.org/abs/2512.16301)
Many papers provide implementation code:
[Code](https://github.com/example/repo)
# Pseudo-code for systematic review
def research_adaptation_strategy(task_domain, method_type):
"""
Navigate to specific adaptation category
Args:
task_domain: e.g., "coding", "theorem proving", "IR"
method_type: "RL", "SFT", "DPO"
Returns:
List of relevant papers with links
"""
# 1. Go to appropriate section (A1, A2, T1, T2)
# 2. Filter by method_type (RL-based vs SFT/DPO)
# 3. Search table for task_domain
# 4. Check paper links, code availability, model backbones
pass
When using this repository for research:
@article{jiang2025adaptation,
title={Adaptation of Agentic AI},
author={Jiang, Pengcheng and Lin, Jiacheng and Shi, Zhiyi and Wang, Zifeng and He, Luxi and Wu, Yichen and Zhong, Ming and Song, Peiyang and Zhang, Qizheng and Wang, Heng and others},
journal={arXiv preprint arXiv:2512.16301},
year={2025}
}
The repository welcomes pull requests for:
Contribution Pattern:
# 1. Fork the repository
# 2. Add paper to appropriate category in README.md
# 3. Follow existing table format:
# | Time | Method | Venue | Task(s) | Tool(s) | Agent Backbone | Tuning |
# Example entry:
# | 2025.XX | YourMethod | Venue<br>🔗 [Paper](link)<br>💻 [Code](link) | Task | Tools | Model | Method |
# 4. Submit pull request
Signal Type:
Training Methods:
Trend: Growing use of GRPO for tool-augmented agents (2025), especially with Qwen2.5 and DeepSeek models
Q: How do I find papers using a specific model like LLaMA3?
grep -i "llama3" README.md
Q: Which papers have open-source code?
# Look for code icon in tables
grep "code_icon.png" README.md
Q: What are the most recent papers?
# Check Time column (sorted by date within categories)
# Most recent: 2025.11 (Orion), 2025.10 (olmOCR2, AlphaProof)
The repository is actively maintained with:
For questions or issues, check: https://github.com/pat-jj/Awesome-Adaptation-of-Agentic-AI/issues