---
title: "DiG-Plan: Mitigating Early Commitment for Tool-Graph Planning via Diffusion Guidance"
authors:
  - "Yansi Li"
  - "Zhuosheng Zhang"
doi: "10.24963/ijcai.2026/649"
url: "https://ijcai.org/proceedings/2026/649"
pdf: "https://ijcai.org/proceedings/2026/0649.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5828-5836"
track: "Main Track"
keywords:
  - "Planning and Scheduling: Search in planning and scheduling"
  - "Planning and Scheduling: Planning algorithms"
  - "Natural Language Processing: Applications"
  - "Natural Language Processing: Tools"
  - "Natural Language Processing: Language models"
---

# DiG-Plan: Mitigating Early Commitment for Tool-Graph Planning via Diffusion Guidance

Yansi Li, Zhuosheng Zhang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5828-5836. DOI: [10.24963/ijcai.2026/649](https://doi.org/10.24963/ijcai.2026/649)

## Abstract

Generating executable tool plans requires selecting appropriate subsets from tool libraries, a combinatorial search problem with an exponentially large solution space. However, we identify a critical misalignment in predominant approaches: standard autoregressive (AR) decoding suffers from early commitment, where initial token choices rigidly constrain the search trajectory. A controlled study shows that masked denoising raises Pass@10 solution coverage from 0.320 to 0.943 over AR sampling under matched compute. Motivated by this, we propose DiG-Plan, a framework that decouples combinatorial exploration from structural refinement. DiG-Plan employs a diffusion-based proposer to generate diverse tool sets via iterative refinement, followed by an AR refiner for dependency prediction. On TaskBench, DiG-Plan improves over AR baselines by a 10% relative margin, with the largest gains on complex compositional tasks; API-Bank results show that the propose-refine-select design remains effective across domains. Code is available at https://github.com/puddingyeah/DiG-Plan.