title={Agentohana: Design unified data and training pipeline for effective agent learning},
author={Zhang, Jianguo and Lan, Tian and Murthy, Rithesh and Liu, Zhiwei and Yao, Weiran and Zhu, Ming and Tan, Juntao and Hoang, Thai and Liu, Zuxin and Yang, Liangwei and others},
journal={arXiv preprint arXiv:2402.15506},
year={2024}
}
@article{chen-etal:fireact,
title={Fireact: Toward language agent fine-tuning},
author={Chen, Baian and Shu, Chang and Shareghi, Ehsan and Collier, Nigel and Narasimhan, Karthik and Yao, Shunyu},
journal={arXiv preprint arXiv:2310.05915},
year={2023}
}
@inproceedings{singh-etal:singhagentic,
title={Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning},
author={Singh, Joykirat and Pandya, Yash and Vajreshwari, Pranav and Magazine, Raghav and Nambi, Akshay},
@@ -27,7 +27,7 @@ An agent's autonomy refers to its ability to independently determine the steps r
\begin{tcolorbox}[frame empty]
\begingroup
% \setlength{\leftskip}{0.3em}
% \setlength{\leftskip}{-0.3em}
% \setlength{\rightskip}{2em}
\centering
\renewcommand{\arraystretch}{1.5}
...
...
@@ -45,6 +45,13 @@ $\cdots \cdots$
\end{tcolorbox}
\begin{figure}
\centering
\caption{111}
\label{fig:agent-planning-sft}
\end{figure}
One commonly used approach to improve the planning capability of LLM-based agents is instruction tuning \citep{chen-etal:fireact,zhang-etal:agentohana}. As illustrated in Figure \ref{fig:agent-planning-sft}, this process typically consists of three stages: (i) preparing environments and planning tasks; (ii) synthesizing expert-level trajectories, which consist of sequences of action-observation pairs, on these tasks; and (iii) instruction-tuning LLMs using the synthesized trajectory data. Specifically, expert trajectories can be generated by leveraging state-of-the-art LLMs as agent policies and selecting high-quality trajectories based on predefined reward signals or task success criteria. Through this process, LLMs can acquire planning behaviors from demonstrations and improve their ability to generate structured execution plans.
