Complete Cyclic Subtask Graphs for Tool-Using LLM Agents: Flexibility, Cost, and Bottlenecks in Long-Horizon Workflows
3.80T1 sourcearXiv cs.MA
Source record
Published by arXiv cs.MA (T1 source). The original is at https://arxiv.org/abs/2604.22820.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryStudies complete cyclic subtask graphs for tool-using LLM agents, comparing Spec-Cyc, Gen-Cyc, DepDAG, and ReAct across TextCraft, ALFWorld, and Finance-Agent. Identifies three task 'signatures'—prerequisite-chain, partially-observable-recovery, and open-ended-evidence-synthesis—each favoring different workflow control patterns. Concludes cyclic graphs serve best as diagnostic tools for judging when backtracking is worth its cost.
Why it mattersProposes a workflow-signature matrix that matches task type to controller style, and quantifies the cost overhead of flexible backtracking. Useful for anyone designing long-horizon agent pipelines.
Cited by
No citations on record.
