UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910253138837504 |
|---|---|
| author | Liang, Yijuan Chen, Xinghao Ge, Yifan Wu, Ziyi Wu, Hao Zeng, Changyu Xing, Wei Shen, Xiaoyu |
| author_facet | Liang, Yijuan Chen, Xinghao Ge, Yifan Wu, Ziyi Wu, Hao Zeng, Changyu Xing, Wei Shen, Xiaoyu |
| contents | Tool-use capability is a fundamental component of LLM agents, enabling them to interact with external systems through structured function calls. However, existing research exhibits inconsistent interaction representations, largely overlooks the structural distribution of tool-use trajectories, and relies on incompatible evaluation benchmarks. We present UniToolCall, a unified framework for tool learning that standardizes the entire pipeline from toolset construction and dataset generation to evaluation. The framework curates a large tool pool of 22k+ tools and constructs a hybrid training corpus of 390k+ instances by combining 10 standardized public datasets with structurally controlled synthetic trajectories. It explicitly models diverse interaction patterns, including single-hop vs. multi-hop and single-turn vs. multi-turn, while capturing both serial and parallel execution structures. To support coherent multi-turn reasoning, we further introduce an Anchor Linkage mechanism that enforces cross-turn dependencies. Furthermore, we convert 7 public benchmarks into a unified Query--Action--Observation--Answer (QAOA) representation with fine-grained evaluation at the function-call, turn, and conversation levels. Experiments show that fine-tuning Qwen3-8B on our dataset substantially improves tool-use performance. Under the distractor-heavy Hybrid-20 setting, achieves 93.0% single-turn Strict Precision, outperforming commercial models including GPT, Gemini, and Claude. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_11557 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents Liang, Yijuan Chen, Xinghao Ge, Yifan Wu, Ziyi Wu, Hao Zeng, Changyu Xing, Wei Shen, Xiaoyu Artificial Intelligence Tool-use capability is a fundamental component of LLM agents, enabling them to interact with external systems through structured function calls. However, existing research exhibits inconsistent interaction representations, largely overlooks the structural distribution of tool-use trajectories, and relies on incompatible evaluation benchmarks. We present UniToolCall, a unified framework for tool learning that standardizes the entire pipeline from toolset construction and dataset generation to evaluation. The framework curates a large tool pool of 22k+ tools and constructs a hybrid training corpus of 390k+ instances by combining 10 standardized public datasets with structurally controlled synthetic trajectories. It explicitly models diverse interaction patterns, including single-hop vs. multi-hop and single-turn vs. multi-turn, while capturing both serial and parallel execution structures. To support coherent multi-turn reasoning, we further introduce an Anchor Linkage mechanism that enforces cross-turn dependencies. Furthermore, we convert 7 public benchmarks into a unified Query--Action--Observation--Answer (QAOA) representation with fine-grained evaluation at the function-call, turn, and conversation levels. Experiments show that fine-tuning Qwen3-8B on our dataset substantially improves tool-use performance. Under the distractor-heavy Hybrid-20 setting, achieves 93.0% single-turn Strict Precision, outperforming commercial models including GPT, Gemini, and Claude. |
| title | UniToolCall: Unifying Tool-Use Representation, Data, and Evaluation for LLM Agents |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2604.11557 |