Topology-Consistent Task Planning over Cellular Workflow Complexes for LLM-based Agents
Quick Answer
TopoPlanner is a new topology-consistent planning framework for LLM agents that enhances task planning by addressing complex workflows like verification-correction loops and merges.
Quick Take
It demonstrates consistent improvements over existing prompt-based and graph-enhanced methods across four benchmarks, showcasing its effectiveness in real-world tool orchestration.
Key Points
- TopoPlanner utilizes cellular workflow complexes for enhanced task planning.
- It addresses complex sub-task dependencies that existing planners struggle with.
- The framework shows improvements in topology-guided workflows over traditional methods.
- Experiments were conducted on four tool-planning benchmarks.
- Results indicate better performance with local backbones.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Task planning for LLM agents requires workflows that satisfy both user intent and complex sub-task dependencies. While existing planners work well for sequential or directed acyclic graph (DAG)-like structures, they struggle with workflow patterns such as verification-correction loops, convergent branch merging, and reusable intermediate states that arise naturally in real-world tool orchestration. We present TopoPlanner, a topology-consistent planning framework that lifts tool dependency graphs into cellular workflow complexes and uses them as topologyaware context for LLM tool planning. TopoPlanner retrieves a request-relevant closed subcomplex through cosheaf-consistent cellular retrieval, performs multidimensional structural reasoning over the retrieved topology, and interfaces the resulting cellular representation with the planner LLM for tool-sequence generation. Experiments on four tool-planning benchmarks with topology-guided loop, merge, and loop-merge workflows show consistent improvements over prompt-based and graph-enhanced baselines across different local LLM backbones.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.07004 [cs.AI] |
| (or arXiv:2610.07004v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07004 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Sen Zhao [view email]
[v1]
Sun, 4 Oct 2026 10:24:20 UTC (3,512 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →HOBA: Hierarchical On-Policy Bidding Agents for Adaptive Online Advertising
HOBA (Hierarchical On-policy Bidding Agents) is a novel hierarchical reinforcement learning framework that enhances online advertising bidding systems by improving adaptability and reducing hyperparameter tuning costs. It utilizes a for hyperparameter inference, a SARSA agent for expert model selection, and a dynamic expert pool for bid execution, achieving a +3.6% increase in target cost during large-scale deployment and outperforming state-of-the-art baselines on AuctionNet.