SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.01622 · ICML · 2024

TravelPlanner: A Benchmark for Real-World Planning with Language Agents

Kai Zhang, Jian Xie, Jiangjie Chen, Yu Su, Tinghui Zhu, ♠ Renze, Lou Yuandong, Tian Yanghua

arXiv · PDF · Open in the Atlas

Code that ran

17 of 17 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 4 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 13 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2402.01622.json · badge: /api/ran/2402.01622.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
OSU-NLP-Group/TravelPlanner canonical (classifier label)
classifier basis: census
13 ran · 0 no recorded run
llv22/travelplanner_forward reimplementation (classifier label)
classifier basis: classifier_2025
4 ran · 0 no recorded run
FunctionStatusWhere it lives
calc_cost_w_tokens Ran
ran, honoured the contract we drafted
OSU-NLP-Group/TravelPlanner/postprocess/openai_request.py
code served (permissive licence) · get_code("3589cdef8c72142f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

# Set seed for reproducibility
torch.manual_seed(0)

# Define arguments for calc_cost_w_tokens
ARGS = {
    'total_tokens': 1000,
    'model_name': 'gpt-3.5-turbo'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
convert_bool_values Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/evaluation/hard_constraint.py
code served (permissive licence) · get_code("144615fb5218a49a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create test data with various nested structures containing numpy booleans
item = {
    "nested_dict": {
        "bool_value": np.bool_(True),
        "int_value": 42,
        "list_with_bool": [np.bool_(False), np.bool_(True), 5],
    },
    "tuple_with_bool": (np.bool_(True), "string", np.bool_(False)),
    "simple_list": [1, 2, 3],
    "another_bool": np.bool_(True),
    "float_value": 3.14,
}

ARGS = {
    "item": item,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
convert_reference_information Ran
ran; our contract draft was wrong, not the code
llv22/travelplanner_forward/tools/planner/sole_planning.py
pointer only (licence: NONE) · get_code("77c486c16345d89e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llv22/travelplanner_forward when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

count_consecutive_values Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
OSU-NLP-Group/TravelPlanner/evaluation/commonsense_constraint.py
code served (permissive licence) · get_code("2f9ab3378b37a898")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 2, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Analyzing the function:
# - Takes a parameter 'lst' which should be a list
# - The function iterates through the list and counts consecutive equal values
# - Returns a list of tuples containing (value, count)
# - Works with any comparable values

# Create test data - a list with consecutive repeated values
test_list = [1, 1, 1, 2, 2, 3, 3, 3, 3, 4, 1, 1]

ARGS = {
    "lst": test_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family int — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_data_display Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/utils/func.py
code served (permissive licence) · get_code("acd03219e0d9155b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

# Define a small CSS content string
css_content = """
.query-highlighted { background-color: #ffffcc; }
.highlighted { background-color: #e6f2ff; }
.highlighted-alt { background-color: #ffe6e6; }
"""

# Define sample data matching the structure expected by the function
data = [
    {
        'query': 'Find a flight from NYC to LA',
        'days': 5,
        'visiting_city_number': 2,
        'date': '2024-01-15',
        'org': 'New York',
        'dest': 'Los Angeles',
        'people_number': 3,
        'budget': 2000,
        'local_constraint': {
            'house rule': 'No smoking',
            'cuisine': 'Italian',
            'room type': 'Double bed',
            'transportation': 'Car rental'
        }
    },
    {
        'query': 'Book a hotel in Paris',
        'days': 3,
        'visiting_city_number': 1,
        'date': '2024-02-20',
        'org': 'London',
        'dest': 'Paris',
        'people_number': 2,
        'budget': 1500,
        'local_constraint': {
            'house rule': 'Quiet hours after 10pm',
            'cuisine': 'French',
            'room type': 'Single bed',
            'transportation': 'Public transit'
        }
    }
]

# Annotation index (must be between 1 and len(data))
annotation_idx = 1

ARGS = {
    'css_content': css_content,
    'data': data,
    'annotation_idx': annotation_idx
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
estimate_budget Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
OSU-NLP-Group/TravelPlanner/utils/budget_estimation.py
code served (permissive licence) · get_code("a96fb9bafb5a310a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

# Set seed for reproducibility
torch.manual_seed(0)

# Create sample data for testing estimate_budget
# The function expects a list/array of numeric values
data = [100, 150, 200, 250, 300, float('nan'), 175]

# The function takes 'mode' as either 'lowest', 'highest', or 'average'
mode = 'average'

# Define ARGS dict with all parameters
ARGS = {
    'data': data,
    'mode': mode
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
extract_before_parenthesis Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/agents/greedy_search.py
code served (permissive licence) · get_code("6f3fd21fac52a0df")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def extract_before_parenthesis(s):
    match = re.search('^(.*?)\\([^)]*\\)', s)
    return match.group(1) if match else s

# The function takes a single string parameter and uses:
# - re.search() to find a pattern
# - match.group(1) to extract the first capturing group
# - Returns the captured group or the original string

ARGS = {
    's': 'function_name(arg1, arg2)'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
extract_from_to Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/evaluation/hard_constraint.py
code served (permissive licence) · get_code("274ece3906f9cef5")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def extract_from_to(text: str):
    """
    Extracts 'A' and 'B' from the format "from A to B" in the given text, with B ending at a comma or the end of the string.
    
    Args:
    - text (str): The input string.
    
    Returns:
    - tuple: A tuple containing 'A' and 'B'. If no match is found, returns (None, None).
    """
    pattern = 'from\\s+(.+?)\\s+to\\s+([^,]+)(?=[,\\s]|$)'
    matches = re.search(pattern, text)
    return matches.groups() if matches else (None, None)


ARGS = {
    "text": "Please go from London to Paris, it's a nice trip"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
extract_numbers_from_filenames Ran
ran, honoured the contract we drafted
llv22/travelplanner_forward/tools/planner/sole_planning.py
pointer only (licence: NONE) · get_code("7877a8bad070567d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llv22/travelplanner_forward when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_query_number Ran
ran, honoured the contract we drafted
OSU-NLP-Group/TravelPlanner/utils/func.py
code served (permissive licence) · get_code("6f08d8cc8543ebda")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re

def extract_query_number(query_string):
    """
    Extract the number from a query string formatted as "Query X" or "Query X --- Done".
    
    Args:
    - query_string (str): The input string.
    
    Returns:
    - int: The extracted number if found, else None.
    """
    pattern = 'Query (\\d+)'
    match = re.search(pattern, query_string)
    return int(match.group(1)) if match else None


ARGS = {
    "query_string": "Query 42"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
format_step Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/agents/tool_agents.py
code served (permissive licence) · get_code("84d28fbaf38b0a59")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

ARGS = {
    "step": "\n  This is a test step with newlines\n  and extra whitespace  \n"
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_city_list Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/agents/greedy_search.py
code served (permissive licence) · get_code("1a12abefb2af139e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from unittest.mock import patch, mock_open
import tempfile
import os

torch.manual_seed(0)

# Mock the file reading since we can't do actual file I/O
mock_city_data = """New York\tNY
Los Angeles\tCA
San Francisco\tCA
Los Vegas\tNV
Phoenix\tAZ"""

# We need to mock the file open to avoid actual file I/O
def mock_file_open(*args, **kwargs):
    return mock_open(read_data=mock_city_data)()

# Patch the open function in the function's module
import builtins
original_open = builtins.open

def patched_open(path, *args, **kwargs):
    if 'citySet_with_states.txt' in path:
        return mock_open(read_data=mock_city_data)()
    return original_open(path, *args, **kwargs)

builtins.open = patched_open

try:
    ARGS = {
        'days': 3,
        'deparure_city': 'New York',
        'destination': 'NY'
    }
finally:
    builtins.open = original_open
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
limited_execution_time Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/postprocess/openai_request.py
code served (permissive licence) · get_code("cf6909a439ea97be")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Define a mock function that simulates an LLM API call
def mock_llm_func(**kwargs):
    return "Generated response"

# Define a mock model (can be a string or object, based on usage)
mock_model = "gpt-3.5-turbo"

# Define a mock prompt (messages for GPT, or prompt string for others)
mock_prompt = [
    {"role": "user", "content": "Hello, how are you?"}
]

# Define temperature
mock_temp = 0.7

# Define max_tokens
mock_max_tokens = 2048

# Define optional kwargs
mock_kwargs = {
    "top_p": 0.9,
    "frequency_penalty": 0.0
}

ARGS = {
    "func": mock_llm_func,
    "model": mock_model,
    "prompt": mock_prompt,
    "temp": mock_temp,
    "max_tokens": mock_max_tokens,
    "default": None,
    "kwargs": mock_kwargs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
func_timeout, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_line_json_data Ran
ran; our contract draft was wrong, not the code
llv22/travelplanner_forward/tools/planner/sole_planning.py
pointer only (licence: NONE) · get_code("959eef29433a0a23")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llv22/travelplanner_forward when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

openai_unit_price Ran
ran, honoured the contract we drafted
OSU-NLP-Group/TravelPlanner/postprocess/openai_request.py
code served (permissive licence) · get_code("db2cf6f145458dbc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

# Set seed for reproducibility
torch.manual_seed(0)

# Define ARGS dictionary with valid arguments for openai_unit_price
ARGS = {
    'model_name': 'gpt-4-turbo',
    'token_type': 'prompt'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
parse_action Ran
ran; our contract draft was wrong, not the code
llv22/travelplanner_forward/agents/tool_agents.py
pointer only (licence: NONE) · get_code("744db5623472fa82")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, llv22/travelplanner_forward when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

transportation_match Ran
ran; our contract draft was wrong, not the code
OSU-NLP-Group/TravelPlanner/evaluation/commonsense_constraint.py
code served (permissive licence) · get_code("610e4862668ffe73")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

# Call torch.manual_seed for reproducibility
torch.manual_seed(0)

# The function takes a single parameter: text (a string)
# It checks for substrings in the lowercased text
# We need to provide a text string that will be processed

ARGS = {
    'text': 'I took a taxi to the airport'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 1 the same month, 3 later) and 13 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
format_step 5 other papers
3 dated earlier, 1 the same month, 1 later
2303.11366 (2023-03)
Reflexion: Language Agents with Verbal Reinforcement Learning
2306.13304 (2023-06)
ToolQA: A Dataset for LLM Question Answering with External Tools
2308.05960 (2023-08)
BOLAA: Benchmarking and Orchestrating LLM-augmented Autonomous Agents
and 2 more not named here
calc_cost_w_tokens 3 other papers
1 dated earlier, 2 later
2305.13300 (2023-05)
Adaptive Chameleon or Stubborn Sloth: REVEALING THE BEHAVIOR OF LARGE LANGUAGE MODELS IN KNOWLEDGE CONFLICTS
2404.03302 (2024-04)
How Easily do Irrelevant Inputs Skew the Responses of Large Language Models?
2410.08414 (2024-10)
Understanding the Interplay between Parametric and Contextual Knowledge for Large Language Models
load_line_json_data 3 other papers
1 dated earlier, 2 later
2305.13300 (2023-05)
Adaptive Chameleon or Stubborn Sloth: REVEALING THE BEHAVIOR OF LARGE LANGUAGE MODELS IN KNOWLEDGE CONFLICTS
2404.03302 (2024-04)
How Easily do Irrelevant Inputs Skew the Responses of Large Language Models?
2410.12409 (2024-10)
Revealing the Barriers of Language Agents in Planning
openai_unit_price 1 other paper
1 dated earlier
2305.13300 (2023-05)
Adaptive Chameleon or Stubborn Sloth: REVEALING THE BEHAVIOR OF LARGE LANGUAGE MODELS IN KNOWLEDGE CONFLICTS

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Planning has been part of the core pursuit for artificial intelligence since its conception, but earlier AI agents mostly focused on constrained settings because many of the cognitive substrates necessary for human-level planning have been lacking. Recently, language agents powered by large language models (LLMs) have shown interesting capabilities such as tool use and reasoning. Are these language agents capable of planning in more complex settings that are out of the reach of prior AI agents? To advance this investigation, we propose TravelPlanner, a new planning benchmark that focuses on travel planning, a common real-world planning scenario. It provides a rich sandbox environment, various tools for accessing nearly four million data records, and 1,225 meticulously curated planning intents and reference plans. Comprehensive evaluations show that the current language agents are not yet capable of handling such complex planning tasks-even GPT-4 only achieves a success rate of 0.6%. Language agents struggle to stay on task, use the right tools to collect information, or keep track of multiple constraints. However, we note that the mere possibility for language agents to tackle such a complex problem is in itself non-trivial progress. TravelPlanner provides a challenging yet meaningful testbed for future language agents.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2402.01622")
get_code_for_paper("2402.01622")
have("2402.01622")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2402.01622.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2402.01622.svg)](https://syntology.ai/paper/2402.01622)

Connect an agent — have() is free.