| TITLE = """<h1 align="center" id="space-title">TravelPlanner Leaderboard</h1>""" | |
| INTRODUCTION_TEXT = """ | |
| TravelPlanner is a benchmark crafted for evaluating language agents in tool-use and complex planning within multiple constraints. (See our [paper](https://arxiv.org/pdf/2402.01622.pdf) for more details.) | |
| ## Data | |
| In TravelPlanner, for a given query, language agents are expected to formulate a comprehensive plan that includes transportation, daily meals, attractions, and accommodation for each day. | |
| For constraints, from the perspective of real world applications, we design three types of them: Environment Constraint, Commonsense Constraint, and Hard Constraint. | |
| TravelPlanner comprises 1,225 queries in total. The number of days and hard constraints are designed to test agents' abilities across both the breadth and depth of complex planning. | |
| TravelPlanner data can be found in [this dataset](https://huggingface.co/datasets/osunlp/TravelPlanner). | |
| ## Submission Guidelines for TravelPlanner | |
| Participants are invited to submit results for both validation and testing phases. The submissions will be evaluated based on several metrics: delivery rate, commonsense constraint pass rate (micro/macro), hard constraint pass rate (micro/macro), and the final pass rate. | |
| ### Format of Submission: | |
| Submissions must be in the form of a JSON-line file. Each line should adhere to the following structure: | |
| ``` | |
| {"idx":0,"query":"Natural Language Query","plan":[{"day": 1, "current_city": "from [City A] to [City B]", "transportation": "Flight Number: XXX, from A to B", "breakfast": "Name, City", "attraction": "Name, City;Name, City;...;Name, City;", "lunch": "Name, City", "dinner": "Name, City", "accommodation": "Name, City"}, {"day": 2, "current_city": "City B", "transportation": "-", "breakfast": "Name, City", "attraction": "Name, City;Name, City;", "lunch": "Name, City", "dinner": "Name, City", "accommodation": "Name, City"}, ...]} | |
| ``` | |
| Explanation of Fields: | |
| #### day: | |
| Description: Indicates the specific day in the itinerary. | |
| Format: Enter the numerical value representing the sequence of the day within the travel plan. For instance, '1' for the first day, '2' for the second day, and so on. | |
| #### current city: | |
| Description: Indicates the city where the traveler is currently located. | |
| Format: When there is a change in location, use "from [City A] to [City B]" to denote the transition. If remaining in the same city, simply use the city's name (e.g., "City A"). | |
| #### transportation: | |
| Description: Specifies the mode of transportation used. | |
| Format: For flights, include the details in the format "Flight Number: XXX, from [City A] to [City B]". For self-driven or taxi travel, use "self-driving/taxi, from [City A] to [City B]". If there is no travel between cities on that day, use "-". | |
| #### breakfast, lunch, and dinner: | |
| Description: Details about dining arrangements. | |
| Format: Use "Name, City" to specify the chosen restaurant and its location. If a meal is not planned, use "-". | |
| #### attraction: | |
| Description: Information about attractions visited. | |
| Format: List attractions as "Name, City". If visiting multiple attractions, separate them with a semicolon ";". If no attraction is planned, use "-". | |
| Please refer to [this](https://huggingface.co/datasets/osunlp/TravelPlanner/resolve/main/example_submission.jsonl?download=true) for example submission file. | |
| Submission made by our team are labelled "TravelPlanner Team". Each submission will be automatically evaluated and scored based on the predefined metrics. You can then obtain the scores and download the detailed constraint pass rates after the evaluation. | |
| ## ⚠️Warnings | |
| We release our evaluation scripts to foster innovation and aid the development of new methods. We encourage the use of evaluation feedback in training set, such as implementing reinforcement learning techniques, to enhance learning. However, we strictly prohibit any form of cheating in the validation and test sets to uphold the fairness and reliability of the benchmark's evaluation process. We reserve the right to disqualify results if we find any of the following violations: | |
| 1. Reverse engineering of our dataset, which includes, but is not limited to: | |
| - Converting our natural language queries in the test set to structured formats (e.g., JSON) for optimization and unauthorized evaluation. | |
| - Deriving data point entries using the hard rules from our data construction process, without accessing the actual database. | |
| - Other similar manipulations. | |
| 2. Hard coding or explicitly writing evaluation cues into prompts by hand, such as direct hints of common sense, which contradicts our goals as it lacks generalizability and is limited to this specific benchmark. | |
| 3. Any other human interference strategies that are tailored specifically to this benchmark but lack generalization capabilities. | |
| ## Show Your Results on Leaderborad | |
| If you are interested in showing your results on our leaderboard, we invite you to reach out to us. Please send an email to [us](mailto:[email protected]) including the following details: evaluation mode, fondation model, tool-use strategy, planning strategy, organization, and your paper link (if available), along with your submission files. | |
| """ | |
| CITATION_BUTTON_LABEL = "Copy the following snippet to cite these results" | |
| CITATION_BUTTON_TEXT = r"""@article{Xie2024TravelPlanner, | |
| author = {Jian Xie, Kai Zhang, Jiangjie Chen, Tinghui Zhu, Renze Lou, Yuandong Tian, Yanghua Xiao, Yu Su}, | |
| title = {TravelPlanner: A Benchmark for Real-World Planning with Language Agents}, | |
| journal = {arXiv preprint arXiv: 2402.01622}, | |
| year = {2024} | |
| }""" | |
| def format_error(msg): | |
| return f"<p style='color: red; font-size: 20px; text-align: center;'>{msg}</p>" | |
| def format_warning(msg): | |
| return f"<p style='color: orange; font-size: 20px; text-align: center;'>{msg}</p>" | |
| def format_log(msg): | |
| return f"<p style='color: green; font-size: 20px; text-align: center;'>{msg}</p>" | |
| def model_hyperlink(link, model_name): | |
| return f'<a target="_blank" href="{link}" style="color: var(--link-text-color); text-decoration: underline;text-decoration-style: dotted;">{model_name}</a>' | |