Divide-Then-Aggregate: An Efficient Tool Learning Method via Parallel Tool Invocation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909621698953216 |
|---|---|
| author | Zhu, Dongsheng Shi, Weixian Shi, Zhengliang Ren, Zhaochun Wang, Shuaiqiang Yan, Lingyong Yin, Dawei |
| author_facet | Zhu, Dongsheng Shi, Weixian Shi, Zhengliang Ren, Zhaochun Wang, Shuaiqiang Yan, Lingyong Yin, Dawei |
| contents | Although current Large Language Models (LLMs) exhibit impressive capabilities, performing complex real-world tasks still requires tool learning. Mainstream methods, such as CoT/ReAct, rely on step-by-step tool invocation to interact with external environments, but they are limited in perceptual scope and lack adequate task-planning capability. To address these limitations, other studies introduce the first Search-based Decision Tree (DFSDT), which still suffers from the high computational cost. In this paper, we introduce a novel parallel tool invocation paradigm, DTA-Llama (Divide-Then-Aggregate Llama). First, we transform traditional tree-based tool search paths into Directed Acyclic Graph (DAG) structure, generating a high-quality parallel tool invocation dataset. The DTA-Llama is then trained on the dataset to learn to iteratively divide the current task into several parallel tool invocation sub-tasks and aggregate the invocation results to decide the next actions. Furthermore, we introduce an efficient inference framework inspired by the Process/Threads mechanism when applying the DTA-Llama to practical tasks. Experimental results show that our approach substantially enhances task performance while reducing token consumption and inference time. Llama2-7B, using our method, is comparable to the official parallel function calling method of GPT-3.5. The relevant code, dataset, and model weights are available at https://corn0205.github.io/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_12432 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Divide-Then-Aggregate: An Efficient Tool Learning Method via Parallel Tool Invocation Zhu, Dongsheng Shi, Weixian Shi, Zhengliang Ren, Zhaochun Wang, Shuaiqiang Yan, Lingyong Yin, Dawei Machine Learning Artificial Intelligence Computation and Language Although current Large Language Models (LLMs) exhibit impressive capabilities, performing complex real-world tasks still requires tool learning. Mainstream methods, such as CoT/ReAct, rely on step-by-step tool invocation to interact with external environments, but they are limited in perceptual scope and lack adequate task-planning capability. To address these limitations, other studies introduce the first Search-based Decision Tree (DFSDT), which still suffers from the high computational cost. In this paper, we introduce a novel parallel tool invocation paradigm, DTA-Llama (Divide-Then-Aggregate Llama). First, we transform traditional tree-based tool search paths into Directed Acyclic Graph (DAG) structure, generating a high-quality parallel tool invocation dataset. The DTA-Llama is then trained on the dataset to learn to iteratively divide the current task into several parallel tool invocation sub-tasks and aggregate the invocation results to decide the next actions. Furthermore, we introduce an efficient inference framework inspired by the Process/Threads mechanism when applying the DTA-Llama to practical tasks. Experimental results show that our approach substantially enhances task performance while reducing token consumption and inference time. Llama2-7B, using our method, is comparable to the official parallel function calling method of GPT-3.5. The relevant code, dataset, and model weights are available at https://corn0205.github.io/ |
| title | Divide-Then-Aggregate: An Efficient Tool Learning Method via Parallel Tool Invocation |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2501.12432 |