dorsal/arxiv
View SchemaiTIMO: An LLM-empowered Synthesis Dataset for Travel Itinerary Modification
| Authors | Zhuoxuan Huang, Yunshan Ma, Hongyu Zhang, Hua Ma, Zhu Sun |
|---|---|
| Categories | |
| ArXiv ID | 2601.10609vv3 |
| URL | https://arxiv.org/abs/2601.10609 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
Addressing itinerary modification is crucial for enhancing the travel experience as it is a frequent requirement during traveling. However, existing research mainly focuses on fixed itinerary planning, leaving modification underexplored. To bridge this gap, we formally define the itinerary modification task and introduce iTIMO, a dataset specifically tailored for this purpose. We identify the lack of {\itshape need-to-modify} itinerary data as the critical bottleneck hindering research on this task and propose a general pipeline to overcome it. This pipeline frames the generation of such data as an intent-driven perturbation task. It instructs large language models to perturb real world itineraries using three atomic editing operations: REPLACE, ADD, and DELETE. Each perturbation is grounded in three intents, including disruptions of popularity, spatial distance, and category diversity. Furthermore, a hybrid evaluation metric is designed to ensure perturbation effectiveness. We conduct comprehensive experiments on iTIMO, revealing the limitations of current LLMs and lead to several valuable directions for future research. Dataset and corresponding code are available at https://github.com/zelo2/iTIMO.
{
"annotation_id": "9b00bdca-3216-46b3-b8cf-9acd2a12e55e",
"date_created": "2026-02-17T05:53:26.420000Z",
"date_modified": "2026-02-17T05:53:26.420000Z",
"file_hash": "2482c47c752573f15a38068761d545d5ae1bf1d03c2e8d37a74473dc46d5ade8",
"private": false,
"record": {
"abstract": "Addressing itinerary modification is crucial for enhancing the travel experience as it is a frequent requirement during traveling. However, existing research mainly focuses on fixed itinerary planning, leaving modification underexplored. To bridge this gap, we formally define the itinerary modification task and introduce iTIMO, a dataset specifically tailored for this purpose. We identify the lack of {\\itshape need-to-modify} itinerary data as the critical bottleneck hindering research on this task and propose a general pipeline to overcome it. This pipeline frames the generation of such data as an intent-driven perturbation task. It instructs large language models to perturb real world itineraries using three atomic editing operations: REPLACE, ADD, and DELETE. Each perturbation is grounded in three intents, including disruptions of popularity, spatial distance, and category diversity. Furthermore, a hybrid evaluation metric is designed to ensure perturbation effectiveness. We conduct comprehensive experiments on iTIMO, revealing the limitations of current LLMs and lead to several valuable directions for future research. Dataset and corresponding code are available at https://github.com/zelo2/iTIMO.",
"arxiv_id": "2601.10609",
"authors": [
"Zhuoxuan Huang",
"Yunshan Ma",
"Hongyu Zhang",
"Hua Ma",
"Zhu Sun"
],
"categories": [
"cs.IR"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "iTIMO: An LLM-empowered Synthesis Dataset for Travel Itinerary Modification",
"url": "https://arxiv.org/abs/2601.10609",
"version": "v3"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "03f0ae30-ff4f-4344-8e25-b8c943b64ec9",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}