Replan, Repair, or Edit? A Unified Empirical Evaluation of Travel Agents for Itinerary Revision under Resource Disruptions
3.00T1 sourcearXiv cs.MA
Source record
Published by arXiv cs.MA (T1 source). The original is at https://arxiv.org/abs/2609.19654.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryEmpirical study comparing three approaches to revising travel itineraries after resource disruptions: LLM-Z3 full replanning, IPyHOPPER hierarchical repair, and iTIMO local editing. Tested on 500 single-disruption and 200 compound-disruption cases. Hierarchical and local repair preserved more accepted commitments than full replanning, with varying computational costs.
Why it mattersQuantifies the trade-off between commitment preservation and feasibility recovery in agent revision workflows, with concrete cost profiles per approach. Useful for agent designers choosing between replanning and repair strategies.
Cited by
No citations on record.
