TaskCraft: Automated Generation of Agentic Tasks

ACL ARR 2025 May Submission7504 Authors

Abstract: Agentic tasks, which require multi-step problem solving with autonomy, tool use, and adaptive reasoning, are becoming increasingly central to the advancement of NLP and AI. However, existing instruction data lacks tool interaction, and current agentic benchmarks rely on costly human annotation, limiting their scalability. We introduce \textsc{TaskCraft}, an automated workflow for generating difficulty-scalable, multi-tool, and verifiable agentic tasks with execution trajectories. TaskCraft expands atomic tasks using width-based and depth-based expansion to create structurally and hierarchically complex challenges. Empirical results show that these tasks improve prompt optimization in the generation workflow and enhance supervised fine-tuning of agentic foundation models. We present a large-scale synthetic dataset of approximately 32,000 tasks with varying difficulty to support future research on agent tuning and evaluation.

Paper Type: Long

Research Area: Generation

Research Area Keywords: agent, generation, LLM, agentic task

Contribution Types: Approaches to low-resource settings, Data resources

Languages Studied: English

Keywords: agent, task extension

Submission Number: 7504