SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.07071 · 2024

LISO: Lidar-only Self-Supervised 3D Object Detection

arXiv · PDF · Open in the Atlas

Code that ran

9 of 9 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 7 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/2403.07071.json · badge: /api/ran/2403.07071.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
baurst/liso canonical (classifier label)
classifier basis: census
9 ran · 0 no recorded run
FunctionStatusWhere it lives
PCA_rectangle Ran
ran; our contract draft was wrong, not the code
baurst/liso/liso/box_fitting/box_fitting.py
code served (permissive licence) · get_code("cbcaa4890fd39e57")
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. 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
from sklearn.decomposition import PCA

# Create a small cluster point cloud for testing
# PCA_rectangle expects cluster_ptc to be a 2D numpy array of shape (n_points, n_features)
# The function will fit PCA with 2 components, so we need at least 2 features
np.random.seed(0)
cluster_ptc = np.random.randn(5, 4)  # 5 points, 4 features

ARGS = {
    "cluster_ptc": cluster_ptc
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
change_dict_keys Ran
ran; our contract draft was wrong, not the code
baurst/liso/liso/datasets/torch_dataset_commons.py
code served (permissive licence) · get_code("7cbfd63356906e16")
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 torch
import numpy as np

torch.manual_seed(0)

# The function takes two parameters:
# 1. obj: a nested structure (dict, list, set, tuple, or primitive)
# 2. convert: a function that converts keys

# Create a simple nested dictionary with various key types
obj = {
    "key_one": 42,
    "key_two": {
        "nested_key": 3.14,
        "another_nested": [1, 2, 3]
    },
    "key_three": (10, 20),
    "key_four": {100, 200, 300}
}

# Define a convert function that transforms keys
# For example, convert strings to uppercase
def convert(k):
    if isinstance(k, str):
        return k.upper()
    return k

ARGS = {
    "obj": obj,
    "convert": convert
}
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 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
closeness_rectangle Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
baurst/liso/liso/box_fitting/box_fitting.py
code served (permissive licence) · get_code("9bb515ae3fee14ac")
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. 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

# The function takes a point cloud (2D array of points) and optional parameters
# cluster_ptc: numpy array of shape (n_points, 2) - 2D point cloud
# delta: float - angle step in degrees (default 5.0)
# d0: float - minimum distance threshold (default 0.01)

# Create a small test point cloud with some 2D points
np.random.seed(0)
cluster_ptc = np.array([
    [0.0, 0.0],
    [1.0, 0.0],
    [1.0, 1.0],
    [0.0, 1.0],
    [0.5, 0.5]
], dtype=np.float64)

ARGS = {
    "cluster_ptc": cluster_ptc,
    "delta": 5.0,
    "d0": 0.01
}
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 tuple — a tuple output has no array shape of its own; none is 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
compute_focal_loss Ran
ran; our contract draft was wrong, not the code
baurst/liso/liso/losses/centerpoint_loss.py
code served (permissive licence) · get_code("a12a92f5c9b2f69a")
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
import torch.nn.functional as F

torch.manual_seed(0)

# Create small tensors for testing
batch_size = 2
spatial_dim = 4

# gt_center_mask: BoolTensor indicating positive locations
gt_center_mask = torch.tensor([
    [True, False, True, False],
    [False, False, True, True]
], dtype=torch.bool)

# groundtruth_probs: FloatTensor with probability values
groundtruth_probs = torch.tensor([
    [0.9, 0.1, 0.8, 0.2],
    [0.3, 0.4, 0.7, 0.6]
], dtype=torch.float32)

# pred_logits: FloatTensor with predicted logits
pred_logits = torch.tensor([
    [2.5, -1.3, 1.8, -0.5],
    [-0.2, 0.1, 2.1, 1.9]
], dtype=torch.float32)

# ignore_where_true_mask: BoolTensor indicating positions to ignore
ignore_where_true_mask = torch.tensor([
    [False, False, False, True],
    [False, True, False, False]
], dtype=torch.bool)

# Scalar parameters
gamma = 2.0
alpha = 0.25

ARGS = {
    'gt_center_mask': gt_center_mask,
    'groundtruth_probs': groundtruth_probs,
    'pred_logits': pred_logits,
    'gamma': gamma,
    'alpha': alpha,
    'ignore_where_true_mask': ignore_where_true_mask
}
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
get_centermaps_downsampling_factor Ran
ran, honoured the contract we drafted
baurst/liso/liso/datasets/torch_dataset_commons.py
code served (permissive licence) · get_code("a67b95c4cfc8e76d")
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 torch

class CenterPointConfig:
    def __init__(self, use_baseline_parameters):
        self.use_baseline_parameters = use_baseline_parameters

class NetworkConfig:
    def __init__(self, use_baseline_parameters):
        self.centerpoint = CenterPointConfig(use_baseline_parameters)

class Config:
    def __init__(self, use_baseline_parameters):
        self.network = NetworkConfig(use_baseline_parameters)

cfg = Config(use_baseline_parameters=False)

ARGS = {
    "cfg": cfg
}
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 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
get_centermaps_output_grid_size Ran
ran, honoured the contract we drafted
baurst/liso/liso/datasets/torch_dataset_commons.py
code served (permissive licence) · get_code("d70d604a3c8767dd")
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 torch
from types import SimpleNamespace

torch.manual_seed(0)

# Build a config object with nested attributes matching the function's expectations
cfg = SimpleNamespace(
    network=SimpleNamespace(
        name='centerpoint',
        centerpoint=SimpleNamespace(
            use_baseline_parameters=True,
            reduce_receptive_field=0
        ),
        transfusion=SimpleNamespace(
            out_size_factor=8
        )
    )
)

# Define output_grid_size as a simple integer
output_grid_size = 256

ARGS = {
    'cfg': cfg,
    'output_grid_size': output_grid_size
}
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 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
minimum_bounding_rectangle Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
baurst/liso/liso/box_fitting/box_fitting.py
code served (permissive licence) · get_code("f0e0262c5390befc")
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. 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
from scipy.spatial import ConvexHull

# Create a simple set of 2D points that form a non-axis-aligned polygon
# Need at least 3 points to form a convex hull
points = np.array([
    [0.0, 0.0],
    [1.0, 0.5],
    [1.5, 2.0],
    [0.5, 2.5],
    [0.0, 1.5]
], dtype=np.float64)

ARGS = {
    "points": points
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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
prob_heatmap_loss Ran
ran; our contract draft was wrong, not the code
baurst/liso/liso/losses/centerpoint_loss.py
code served (permissive licence) · get_code("faef1b767afe5367")
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
import torch.nn as nn

torch.manual_seed(0)

class LossCfg:
    def __init__(self):
        self.supervised = SupervisedCfg()

class SupervisedCfg:
    def __init__(self):
        self.centermaps = CentermapsCfg()

class CentermapsCfg:
    def __init__(self):
        self.confidence_target = 'gaussian'

loss_cfg = LossCfg()

gt_center_mask = torch.tensor([
    [True, False, True, False],
    [False, True, False, True]
], dtype=torch.bool)

groundtruth_probs = torch.tensor([
    [0.9, 0.1, 0.8, 0.2],
    [0.3, 0.7, 0.4, 0.6]
], dtype=torch.float32)

pred_logits = torch.tensor([
    [2.0, -1.5, 1.8, -0.5],
    [-0.8, 2.2, -1.2, 1.5]
], dtype=torch.float32)

ignore_where_true_mask = torch.tensor([
    [False, False, False, False],
    [False, False, True, False]
], dtype=torch.bool)

ARGS = {
    'loss_cfg': loss_cfg,
    'gt_center_mask': gt_center_mask,
    'groundtruth_probs': groundtruth_probs,
    'pred_logits': pred_logits,
    'ignore_where_true_mask': ignore_where_true_mask
}
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
to_positive_angle Ran
ran; our contract draft was wrong, not the code
baurst/liso/liso/losses/centerpoint_loss.py
code served (permissive licence) · get_code("50451d5311de7a4c")
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 2, dtype family float, shape (2, 4) 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 small tensor with angles (both positive and negative values)
# to test the function's behavior
angle_tensor = torch.tensor([
    [-1.5, 0.5, 3.14, 6.28],
    [-0.5, 1.57, -3.14, 2.0]
], dtype=torch.float32)

ARGS = {
    "angle_rad": angle_tensor
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 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 9 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

3D object detection is one of the most important components in any Self-Driving stack, but current state-of-the-art (SOTA) lidar object detectors require costly & slow manual annotation of 3D bounding boxes to perform well. Recently, several methods emerged to generate pseudo ground truth without human supervision, however, all of these methods have various drawbacks: Some methods require sensor rigs with full camera coverage and accurate calibration, partly supplemented by an auxiliary optical flow engine. Others require expensive high-precision localization to find objects that disappeared over multiple drives. We introduce a novel self-supervised method to train SOTA lidar object detection networks which works on unlabeled sequences of lidar point clouds only, which we call trajectory-regularized self-training. It utilizes a SOTA self-supervised lidar scene flow network under the hood to generate, track, and iteratively refine pseudo ground truth. We demonstrate the effectiveness of our approach for multiple SOTA object detection networks across multiple real-world datasets. Code will be released.

For agents

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

get_harvested_code_for_paper("2403.07071")
get_code_for_paper("2403.07071")
have("2403.07071")

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

curl https://syntology.ai/api/ran/2403.07071.json

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

[![Syntology run record](https://syntology.ai/api/ran/2403.07071.svg)](https://syntology.ai/paper/2403.07071)

Connect an agent — have() is free.