Tool Learning with Foundation Models

By Yujia Qin, Shengding Hu, Yankai Lin, Weize Chen, Ning Ding, Ganqu Cui, Zheni Zeng, Yufei Huang, Chaojun Xiao, Chi Han and 31 more. ACM Computing Surveys, 2023

420 citations recorded by the source corpus.

Humans possess an extraordinary ability to create and utilize tools. With the advent of foundation models, artificial intelligence systems have the potential to be equally adept in tool use as humans. This paradigm, which is dubbed as tool learning with foundation models, combines the strengths of tools and foundation models to achieve enhanced accuracy, efficiency, and automation in problem-solving. This article presents a systematic investigation and comprehensive review of tool learning. We first introduce the background of tool learning, including its cognitive origins, the paradigm shift of foundation models, and the complementary roles of tools and models. Then we recapitulate existing tool learning research and formulate a general framework: starting from understanding the user instruction, models should learn to decompose a complex task into several subtasks, dynamically adjust their plan through reasoning, and effectively conquer each sub-task by selecting appropriate tools. We also discuss how to train models for improved tool-use capabilities and facilitate generalization in tool learning. Finally, we discuss several open problems that require further investigation, such as ensuring trustworthy tool use, enabling tool creation with foundation models, and addressing personalization challenges. Overall, we hope this article could inspire future research in integrating tools with foundation models.

Subfields: Artificial Intelligence, Control and Systems Engineering, Information Systems

View Paper: https://www.semanticscholar.org/paper/352420ee61a8da783ca7750170793613b18b8d9c

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Zhiyuan Liu
2. Maosong Sun
3. Peng Gao
4. Hongsheng Li
5. Wayne Xin Zhao
6. Yu Qiao
7. Yankai Lin
8. Ji-rong Wen

In Control and Systems Engineering

1. Yu Qiao
2. Enhong Chen
3. Hongsheng Li
4. Jean Kaddour

In Information Systems

1. Ji-rong Wen
2. Wayne Xin Zhao
3. Maosong Sun
4. Zhiyuan Liu
5. Xingyao Wang
6. Huan Sun
7. Zhewei Wei
8. W. Zhang

Citations in the corpus

• A Survey on the Memory Mechanism of Large Language Model-based Agents (2024, neutral): “In addition, cognitive psychology has been studied for a long time, so many effective human memory theories and architectures have been accumulated, which can support more advanced capabilities of the agents [147].”
• BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions (2024, methodology): “Tool Statistics In this work, we highlight any libraries and function calls from these libraries as program-based tools for LLMs [50].”
• Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflicts (2023, methodology): “Tool 1 (Schick et al., 2023; Qin et al., 2023) or retrieval augmentation (Mallen et al., 2022; Shi et al., 2023b; Ram et al., 2023) has emerged as a promising solution by providing external information as new evidence to LLMs, such as ChatGPT Plugins and New Bing.”
• Understanding the planning of LLM agents: A survey (2024, methodology): “…to summarize techniques for LLMs [Zhao et al. , 2023a], LLMs for decision-making [Yang et al. , 2023a], reasoning [Sun et al. , 2023], tool learning [Qin et al. , 2023], and autonomous agents [Wang et al. , 2023a], they often lack a detailed analysis of planning ability within the literature.”
• When large language models meet personalization: perspectives of challenges and opportunities (2023, neutral): “Tool learning is an emerging research field that aims to enhance task-solving capabilities by combining specialized tools with foundational models, which has been understood by [216] as two perspectives:”
• The rise and potential of large language model based agents: a survey (2025, positive): “Previous studies showed that LLM-based agents not only have the ability to generalize the use of existing tools, but can also make new tools they need by generating executable programs, or integrating existing tools into more powerful ones [75,241,242].”
• UltraFeedback: Boosting Language Models with High-quality Feedback (2023, positive): “With more powerful backbone models, we believe ULTRAFEEDBACK will greatly benefit autonomous agents (Park et al., 2023; Qin et al., 2023; Qian et al., 2023) and feedback learning (Yao et al.”
• Mind2Web: Towards a Generalist Agent for the Web (2023, background): “On the other hand, such an agent may also turn the entire web into an unprecedentedly powerful and versatile tool [19, 23] that can enhance large language models (LLMs).”
• RepoAgent: An LLM-Powered Open-Source Framework for Repository-level Code Documentation Generation (2024, positive): “As seen in the table, the GPT series significantly outperforms the LLaMA series in parameter identification, and gpt-4-0125 performs the best.”
• Executable Code Actions Elicit Better LLM Agents (2024, background): “The action modules are key to facilitating LLM agents to effectively interact with external entities, including humans (Lee et al., 2022) and tools (Qin et al., 2023a) in the environment (Wang et al., 2023e; Yang et al., 2024a).”
• A survey on large language model based autonomous agents (2023, positive): “BMTools [153] is an open-source repository that extends LLMs with tools and provides a platform for community-driven tool building and sharing.”