SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.03633 · 2024

Neural Graph Map: Dense Mapping with Efficient Loop Closure Integration

arXiv · PDF · Open in the Atlas

Code that ran

8 of 11 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 8 failures of our instrument; 3 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/2405.03633.json · badge: /api/ran/2405.03633.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
kth-rpl/neural_graph_mapping canonical (classifier label)
classifier basis: census
8 ran · 3 no recorded run
FunctionStatusWhere it lives
complex_apply Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/models.py
code served (permissive licence) · get_code("bca12539c3228964")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) 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

torch.manual_seed(0)

# Create complex numbers tensor with shape (..., 2)
# Real part first, imaginary part second
# Shape: (2, 3, 2) - batch of 2, 3 complex numbers each
comp = torch.randn(2, 3, 2)

# Create 2D points tensor with shape (..., 2)
# Shape: (2, 3, 2) - batch of 2, 3 points each
point = torch.randn(2, 3, 2)

ARGS = {
    "comp": comp,
    "point": point,
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) — shape (2, 3, 2) 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
complex_invert Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/models.py
code served (permissive licence) · get_code("221500418b90d60a")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) 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

torch.manual_seed(0)

# Create a complex number tensor with shape (..., 2)
# The last dimension has size 2: [real_part, imaginary_part]
# Use shape (2, 3, 2) to have multiple complex numbers
comp = torch.randn(2, 3, 2)

ARGS = {
    "comp": comp,
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) — shape (2, 3, 2) 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
complex_raw_multiply Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/models.py
code served (permissive licence) · get_code("53a86c9b7da3784e")
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) 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

torch.manual_seed(0)

# Create two complex number tensors
# Shape (..., 2) where last dimension is [real, imaginary]
a = torch.randn(2, 3, 2)  # batch size 2, 3 complex numbers, 2 components each
b = torch.randn(2, 3, 2)  # same shape for broadcasting

ARGS = {
    "a": a,
    "b": b,
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 2) — shape (2, 3, 2) 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
create_layer Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/utils.py
code served (permissive licence) · get_code("605e0c0d4ffa5eae")
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 Sequential. 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

torch.manual_seed(0)

# Define arguments for create_layer
ARGS = {
    "in_features": 4,
    "out_features": 3,
    "activation": torch.nn.ReLU(),
    "bias": True,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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_neighbors Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/graph.py
code served (permissive licence) · get_code("f514004255342669")
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 set. 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.

# Build a simple graph as an adjacency list (dict mapping vertices to their neighbors)
graph = {
    'A': ['B', 'C'],
    'B': ['A', 'D'],
    'C': ['A', 'E'],
    'D': ['B', 'F'],
    'E': ['C'],
    'F': ['D']
}

query_vertices = {'A', 'B'}
max_edges = 2
include_queries = False

ARGS = {
    'graph': graph,
    'query_vertices': query_vertices,
    'max_edges': max_edges,
    'include_queries': include_queries
}
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 set — a set output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a set 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
photometric_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/losses.py
code served (permissive licence) · get_code("88e2ba8898cc53f4")
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 Tensor, 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

torch.manual_seed(0)

# Create small tensors with distinct sizes
# Shape (..., 3) for color tensors - using batch of 2, spatial dim 4
measured_colors = torch.randn(2, 4, 3)
rendered_colors = torch.randn(2, 4, 3)
rendered_color_vars = torch.abs(torch.randn(2, 4, 3)) + 0.1  # Ensure positive variances

ARGS = {
    'mode': 'l2',
    'measured_colors': measured_colors,
    'rendered_colors': rendered_colors,
    'rendered_color_vars': rendered_color_vars
}
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 Tensor, 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
psnr Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/evaluation.py
code served (permissive licence) · get_code("1f49114640bf9b77")
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 torchmetrics.functional as tmf

torch.manual_seed(0)

# Create prediction and target tensors
# The function expects tensors of shape (H, W, C) before permutation
# After operations: permute(2, 0, 1) gives (C, H, W), then unsqueeze(0) gives (1, C, H, W)
# Using distinct dimensions: H=4, W=3, C=2
prediction = torch.rand(4, 3, 2)
target = torch.rand(4, 3, 2)

crop = 0

ARGS = {
    "prediction": prediction,
    "target": target,
    "crop": crop
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchmetrics — 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
remove_vertex Ran
ran; our contract draft was wrong, not the code
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/graph.py
code served (permissive licence) · get_code("612a1a28f85fb84b")
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 copy

# Construct a graph as a dictionary where keys are vertices (integers)
# and values are lists of adjacent vertices
graph = {
    0: [1, 2],
    1: [0, 3],
    2: [0, 3],
    3: [1, 2]
}

vertex = 1

ARGS = {
    "graph": graph,
    "vertex": vertex
}
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 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
benchmark Not yet run
unverified (harvested, no recorded run)
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/utils.py
code served (permissive licence) · get_code("2554d294968c8675")
How it ran

no invocation recorded.

eikonal_term Not yet run
unverified (harvested, no recorded run)
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/losses.py
code served (permissive licence) · get_code("d7b6fb6bb5369cfc")
How it ran

no invocation recorded.

ssim Not yet run
unverified (harvested, no recorded run)
kth-rpl/neural_graph_mapping/src/neural_graph_mapping/evaluation.py
code served (permissive licence) · get_code("995b83459084d517")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 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 11 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

Neural field-based SLAM methods typically employ a single, monolithic field as their scene representation. This prevents efficient incorporation of loop closure constraints and limits scalability. To address these shortcomings, we propose a novel RGB-D neural mapping framework in which the scene is represented by a collection of lightweight neural fields which are dynamically anchored to the pose graph of a sparse visual SLAM system. Our approach shows the ability to integrate large-scale loop closures, while requiring only minimal reintegration. Furthermore, we verify the scalability of our approach by demonstrating successful building-scale mapping taking multiple loop closures into account during the optimization, and show that our method outperforms existing state-of-the-art approaches on large scenes in terms of quality and runtime. Our code is available open-source at https://github.com/KTH-RPL/neural_graph_mapping.

For agents

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

get_harvested_code_for_paper("2405.03633")
get_code_for_paper("2405.03633")
have("2405.03633")

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

curl https://syntology.ai/api/ran/2405.03633.json

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

[![Syntology run record](https://syntology.ai/api/ran/2405.03633.svg)](https://syntology.ai/paper/2405.03633)

Connect an agent — have() is free.