SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.02511 · 2024

LLM-A*: Large Language Model Enhanced Incremental Heuristic Search on Path Planning

arXiv · PDF · Open in the Atlas

Code that ran

6 of 6 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; 3 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 1 failure 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/2407.02511.json · badge: /api/ran/2407.02511.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
SilinMeng0510/llm-astar canonical (classifier label)
classifier basis: census
6 ran · 0 no recorded run
FunctionStatusWhere it lives
filter_collision_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SilinMeng0510/llm-astar/llmastar/utils/filter.py
code served (permissive licence) · get_code("1b5155b0b089faae")
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 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 math

def point_circle(px, py, cx, cy, r):
    distX = px - cx
    distY = py - cy
    distance = math.sqrt(distX ** 2 + distY ** 2)
    return distance <= r

def line_point(x1, y1, x2, y2, px, py):
    d1 = math.dist((px, py), (x1, y1))
    d2 = math.dist((px, py), (x2, y2))
    lineLen = math.dist((x1, y1), (x2, y2))
    buffer = 0.1
    return abs(d1 + d2 - lineLen) <= buffer

def line_circle(x1, y1, x2, y2, cx, cy, r):
    inside1 = point_circle(x1, y1, cx, cy, r)
    inside2 = point_circle(x2, y2, cx, cy, r)
    if inside1 or inside2:
        return True
    distX = x1 - x2
    distY = y1 - y2
    len = math.sqrt(distX ** 2 + distY ** 2)
    dot = ((cx - x1) * (x2 - x1) + (cy - y1) * (y2 - y1)) / len ** 2
    closestX = x1 + dot * (x2 - x1)
    closestY = y1 + dot * (y2 - y1)
    on_segment = line_point(x1, y1, x2, y2, closestX, closestY)
    if not on_segment:
        return False
    distX = closestX - cx
    distY = closestY - cy
    distance = math.sqrt(distX ** 2 + distY ** 2)
    return distance <= r

def filter_collision_path(start, ends, object):
    new = []
    for end in ends:
        if start[0] == end[0] and start[1] == end[1]:
            continue
        if not line_circle(start[0], start[1], end[0], end[1], object[0], object[1], object[2]):
            new.append(end)
    return new

start = (0.0, 0.0)
ends = [(1.0, 1.0), (2.0, 2.0), (0.0, 0.0), (3.0, 0.5)]
object = (5.0, 5.0, 2.0)

ARGS = {
    "start": start,
    "ends": ends,
    "object": object
}
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 list, ndim 2, 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
line_circle Ran
ran, violated the contract we drafted
SilinMeng0510/llm-astar/llmastar/utils/filter.py
code served (permissive licence) · get_code("6c78dde4a6590ded")
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 bool, 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 math

def point_circle(px, py, cx, cy, r):
    distX = px - cx
    distY = py - cy
    distance = math.sqrt(distX ** 2 + distY ** 2)
    return distance <= r

def line_point(x1, y1, x2, y2, px, py):
    d1 = math.dist((px, py), (x1, y1))
    d2 = math.dist((px, py), (x2, y2))
    lineLen = math.dist((x1, y1), (x2, y2))
    buffer = 0.1
    return abs(d1 + d2 - lineLen) <= buffer

def line_circle(x1, y1, x2, y2, cx, cy, r):
    inside1 = point_circle(x1, y1, cx, cy, r)
    inside2 = point_circle(x2, y2, cx, cy, r)
    if inside1 or inside2:
        return True
    distX = x1 - x2
    distY = y1 - y2
    len = math.sqrt(distX ** 2 + distY ** 2)
    dot = ((cx - x1) * (x2 - x1) + (cy - y1) * (y2 - y1)) / len ** 2
    closestX = x1 + dot * (x2 - x1)
    closestY = y1 + dot * (y2 - y1)
    on_segment = line_point(x1, y1, x2, y2, closestX, closestY)
    if not on_segment:
        return False
    distX = closestX - cx
    distY = closestY - cy
    distance = math.sqrt(distX ** 2 + distY ** 2)
    return distance <= r

ARGS = {
    "x1": 0.0,
    "y1": 0.0,
    "x2": 10.0,
    "y2": 10.0,
    "cx": 5.0,
    "cy": 5.0,
    "r": 2.0,
}
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 bool, 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
list_parse Ran
ran, honoured the contract we drafted
SilinMeng0510/llm-astar/llmastar/utils/utils.py
code served (permissive licence) · get_code("ae5529e4c3e3ea6f")
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 re
import ast

def list_parse(text):
    lists = re.findall('\\[\\[.*?\\]\\]', text, re.DOTALL)
    if lists:
        last_list_str = lists[-1]
        evaluated_list_str = re.sub('(\\d+ - \\d+\\.\\d+)', lambda x: str(eval(x.group(1))), last_list_str)
        try:
            evaluated_list = ast.literal_eval(evaluated_list_str)
            return evaluated_list
        except Exception as e:
            return f'Error evaluating list: {e}'
    else:
        return [[0, 0], [0, 0]]

# Test with a simple text containing nested lists
text_with_lists = "Here is some data: [[1, 2], [3, 4]] and more text"

ARGS = {
    'text': text_with_lists
}
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 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
parse Ran
ran, honoured the contract we drafted
SilinMeng0510/llm-astar/llmastar/utils/utils.py
code served (permissive licence) · get_code("e4695eb63c674f80")
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 re
import torch
import torch.nn as nn

torch.manual_seed(0)

# The parse function takes a command string and extracts numbers from it
# using regex, then converts them to int or float based on whether they contain a decimal point

# Create a simple command string with various numbers
command = "process -3.14 data 42 -100 2.71828 config 0"

ARGS = {
    "command": command
}
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 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
parse_search Ran
ran, honoured the contract we drafted
SilinMeng0510/llm-astar/llmastar/utils/utils.py
code served (permissive licence) · get_code("07de23c8e65814a8")
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 re

def parse_search(command):
    sample = re.findall('\\[(-?\\d+),\\s*(-?\\d+)\\]', command)
    return [(int(x), int(y)) for x, y in sample]

# The function takes a command string and uses regex to find patterns like [num, num]
# It extracts pairs of integers (positive or negative) and returns them as a list of tuples

ARGS = {
    "command": "[10, 20] some text [-5, 15] more text [0, -3]"
}
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 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
point_circle Ran
ran, violated the contract we drafted
SilinMeng0510/llm-astar/llmastar/utils/filter.py
code served (permissive licence) · get_code("273b786cb316ba47")
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 bool, 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 math

# The function point_circle takes 5 scalar parameters representing coordinates
# px, py: point coordinates
# cx, cy: circle center coordinates
# r: circle radius

ARGS = {
    "px": 5.0,
    "py": 3.0,
    "cx": 5.0,
    "cy": 3.0,
    "r": 2.5
}
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 bool, 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

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 6 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.

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

Path planning is a fundamental scientific problem in robotics and autonomous navigation, requiring the derivation of efficient routes from starting to destination points while avoiding obstacles. Traditional algorithms like A* and its variants are capable of ensuring path validity but suffer from significant computational and memory inefficiencies as the state space grows. Conversely, large language models (LLMs) excel in broader environmental analysis through contextual understanding, providing global insights into environments. However, they fall short in detailed spatial and temporal reasoning, often leading to invalid or inefficient routes. In this work, we propose LLM-A*, an new LLM based route planning method that synergistically combines the precise pathfinding capabilities of A* with the global reasoning capability of LLMs. This hybrid approach aims to enhance pathfinding efficiency in terms of time and space complexity while maintaining the integrity of path validity, especially in large-scale scenarios. By integrating the strengths of both methodologies, LLM-A* addresses the computational and memory limitations of conventional algorithms without compromising on the validity required for effective pathfinding.

For agents

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

get_harvested_code_for_paper("2407.02511")
get_code_for_paper("2407.02511")
have("2407.02511")

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

curl https://syntology.ai/api/ran/2407.02511.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.02511.svg)](https://syntology.ai/paper/2407.02511)

Connect an agent — have() is free.