SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2305.19201 · NeurIPS · 2023

DäRF: Boosting Radiance Fields from Sparse Inputs with Monocular Depth Adaptation

Seungryong Kim, Min-Seop Kwak, Jiuhn Song, Seonghoon Park, Honggyu An, Seokju Cho, Sungjin Cho

arXiv · PDF · Open in the Atlas

Code that ran

10 of 10 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, 1 violated it and 3 have no contract verdict; 4 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/2305.19201.json · badge: /api/ran/2305.19201.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
KU-CVLAB/DaRF canonical (classifier label)
classifier basis: census
10 ran · 0 no recorded run
FunctionStatusWhere it lives
DiffAugment Ran
ran, violated the contract we drafted
KU-CVLAB/DaRF/plenoxels/models/discriminator.py
code served (permissive licence) · get_code("2e499d784aa6bc3c")
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 4, dtype family float, shape (2, 3, 4, 5) 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

torch.manual_seed(0)

# Create a small image tensor: batch_size=2, channels=3, height=4, width=5
x = torch.randn(2, 3, 4, 5, dtype=torch.float32)

ARGS = {
    'x': x,
    'policy': 'color,cutout',
    'channels_first': True
}
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 Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
actvn Ran
ran; our contract draft was wrong, not the code
KU-CVLAB/DaRF/plenoxels/models/discriminator.py
code served (permissive licence) · get_code("51e3b2175da36d6c")
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
import torch.nn.functional as F

torch.manual_seed(0)

# The function actvn takes a tensor x and applies leaky_relu with negative slope 0.2
# It needs a tensor as input
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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
apply_min_size Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
KU-CVLAB/DaRF/plenoxels/models/transforms.py
code served (permissive licence) · get_code("8d4e5ab503856d09")
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 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 numpy as np
import cv2
import math

# Create a sample dictionary with required keys
sample = {
    'disparity': np.random.rand(10, 15).astype(np.float32),
    'image': np.random.randint(0, 256, (10, 15, 3), dtype=np.uint8),
    'mask': np.random.rand(10, 15) > 0.5
}

# Define the size parameter
size = (20, 25)

# Define the image interpolation method
image_interpolation_method = cv2.INTER_AREA

ARGS = {
    'sample': sample,
    'size': size,
    'image_interpolation_method': image_interpolation_method
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
cv2, numpy — 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 int — 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
attn_cosine_sim Ran
ran; our contract draft was wrong, not the code
KU-CVLAB/DaRF/plenoxels/models/extractor.py
pointer only (licence: MIT); the stored copy get_code serves, from omerbt/Splice, is pointer only (licence: NONE) · get_code("e9f612ce15936ed2")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from omerbt/Splice, and GitHub detected no licence in that repository 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.

get_normalized_directions Ran
ran, honoured the contract we drafted
KU-CVLAB/DaRF/plenoxels/models/kplane_field.py
code served (permissive licence) · get_code("adc28f78a3e2f5c8")
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)

# The function takes directions and normalizes them from [-1, 1] to [0, 1]
# directions should be a tensor of shape (batch_size, num_directions)
directions = torch.randn(2, 4)  # batch size 2, 4 direction components

ARGS = {
    "directions": directions
}
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
init_grid_param Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
KU-CVLAB/DaRF/plenoxels/models/kplane_field.py
code served (permissive licence) · get_code("18f6227e830af980")
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 ParameterList. 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)

ARGS = {
    'grid_nd': 2,
    'in_dim': 4,
    'out_dim': 3,
    'reso': [8, 6, 5, 7],
    'a': 0.1,
    'b': 0.5
}
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 ParameterList — a ParameterList 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
load_render_pose Ran
ran on a synthesized input
KU-CVLAB/DaRF/plenoxels/datasets/scannet_dataset.py
code served (permissive licence) · get_code("dd074522d02e6097")
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 json
import os
import tempfile
import numpy as np

# Create a temporary directory with the required transforms_video.json file
temp_dir = tempfile.mkdtemp()
subject_dir = os.path.join(temp_dir, '0781_test')
os.makedirs(subject_dir, exist_ok=True)

# Create a valid transforms_video.json file
transforms_data = {
    'frames': [
        {
            'transform_matrix': [[1, 0, 0, 0], [0, 1, 0, 0], [0, 0, 1, 0], [0, 0, 0, 1]],
            'fx': 500.0,
            'fy': 500.0,
            'cx': 256.0,
            'cy': 256.0
        },
        {
            'transform_matrix': [[1, 0, 0, 0.1], [0, 1, 0, 0.2], [0, 0, 1, 0.3], [0, 0, 0, 1]],
            'fx': 500.0,
            'fy': 500.0,
            'cx': 256.0,
            'cy': 256.0
        },
        {
            'transform_matrix': [[1, 0, 0, 0.2], [0, 1, 0, 0.4], [0, 0, 1, 0.6], [0, 0, 0, 1]],
            'fx': 500.0,
            'fy': 500.0,
            'cx': 256.0,
            'cy': 256.0
        }
    ]
}

with open(os.path.join(subject_dir, 'transforms_video.json'), 'w') as fp:
    json.dump(transforms_data, fp)

ARGS = {
    'subject_dir': subject_dir,
    'split': 'train',
    'fewshot': None,
    'load_depth': 1
}
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
load_scannet_images Ran
ran on a synthesized input
KU-CVLAB/DaRF/plenoxels/datasets/scannet_dataset.py
code served (permissive licence) · get_code("e10dc89a2ca0f09f")
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 json
import os
import numpy as np
import torch
import tempfile
import shutil

torch.manual_seed(0)

# Create a temporary directory structure for the test
temp_dir = tempfile.mkdtemp()
subject_dir = os.path.join(temp_dir, 'scene_0781')
os.makedirs(subject_dir)

# Create subdirectories for images and depth maps
img_subdir = os.path.join(subject_dir, 'rgb')
os.makedirs(img_subdir)
depth_subdir = os.path.join(subject_dir, 'rgb', 'midas_depth')
os.makedirs(depth_subdir)
colmap_subdir = os.path.join(subject_dir, 'rgb', 'depth')
os.makedirs(colmap_subdir)
target_depth_subdir = os.path.join(subject_dir, 'rgb', 'target_depth')
os.makedirs(target_depth_subdir)

# Create dummy image files (RGBA format, 64x64)
for i in range(5):
    img_array = np.random.randint(0, 256, (64, 64, 4), dtype=np.uint8)
    img_path = os.path.join(img_subdir, f'{i:06d}.png')
    from PIL import Image
    Image.fromarray(img_array).save(img_path)

# Create dummy depth files for test split
for i in range(5):
    depth_array = np.random.randint(0, 10000, (64, 64), dtype=np.uint16)
    depth_path = os.path.join(target_depth_subdir, f'{i:06d}.png')
    Image.fromarray(depth_array).save(depth_path)
    
    colmap_array = np.random.randint(0, 10000, (64, 64), dtype=np.uint16)
    colmap_path = os.path.join(colmap_subdir, f'{i:06d}.png')
    Image.fromarray(colmap_array).save(colmap_path)

# Create dummy depth files for train split
for i in range(5):
    depth_array = np.random.randint(0, 10000, (64, 64), dtype=np.uint16)
    depth_path = os.path.join(depth_subdir, f'{i:06d}_depth.png')
    Image.fromarray(depth_array).save(depth_path)

# Create transforms_train.json
train_frames = []
for i in range(5):
    frame = {
        'file_path': f'rgb/{i:06d}.png',
        'transform_matrix': np.eye(4).tolist(),
        'fx': 500.0,
        'fy': 500.0,
        'cx': 32.0,
        'cy': 32.0
    }
    train_frames.append(frame)

train_meta = {'frames': train_frames}
with open(os.path.join(subject_dir, 'transforms_train.json'), 'w') as f:
    json.dump(train_meta, f)

# Create transforms_test.json
test_frames = []
for i in range(5):
    frame = {
        'file_path': f'rgb/{i:06d}.png',
        'transform_matrix': np.eye(4).tolist(),
        'fx': 500.0,
        'fy': 500.0,
        'cx': 32.0,
        'cy': 32.0
    }
    test_frames.append(frame)

test_meta = {'frames': test_frames}
with open(os.path.join(subject_dir, 'transforms_test.json'), 'w') as f:
    json.dump(test_meta, f)

ARGS = {
    'subject_dir': subject_dir,
    'split': 'train',
    'fewshot': None,
    'load_depth': 1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, imageio, numpy, torch — 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
normalize_aabb Ran
ran, honoured the contract we drafted
KU-CVLAB/DaRF/plenoxels/models/kplane_field.py
code served (permissive licence) · get_code("9d244f95cc9869be")
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)

# pts: a tensor of points to be normalized
# The function performs element-wise operations: subtraction, multiplication, subtraction
# So pts should be a tensor with numeric values
pts = torch.randn(2, 4)

# aabb: axis-aligned bounding box, used as aabb[0] (min) and aabb[1] (max)
# The function indexes it with [0] and [1], so it needs at least 2 elements
# Each element should be a tensor or array-like with same shape as pts for broadcasting
aabb_min = torch.tensor([-1.0, -1.0, -1.0, -1.0])
aabb_max = torch.tensor([1.0, 1.0, 1.0, 1.0])
aabb = [aabb_min, aabb_max]

ARGS = {
    "pts": pts,
    "aabb": aabb,
}
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
reparametrize Ran
ran on a synthesized input
KU-CVLAB/DaRF/plenoxels/models/discriminator.py
code served (permissive licence) · get_code("d3daffe86c3a152c")
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 small tensors for mu and logstd
# mu: mean of the distribution, shape (2, 4) for batch_size=2, latent_dim=4
mu = torch.randn(2, 4)

# logstd: log standard deviation, same shape as mu
logstd = torch.randn(2, 4)

ARGS = {
    "mu": mu,
    "logstd": logstd,
}
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 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 8 are also present in other papers whose code we hold (5 with at least one paper dated earlier, 0 the same month, 6 later) and 2 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
actvn 6 other papers
5 dated earlier, 1 later
1810.00821 (2018-10)
Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow
1912.05237 (2019-12)
Towards Unsupervised Learning of Generative Models for 3D Controllable Image Synthesis
2002.11810 (2020-02)
On Leveraging Pretrained GANs for Generation with Limited Data
and 3 more not named here
apply_min_size 4 other papers
2 dated earlier, 2 later
1907.01341 (2019-07)
Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer
2210.01035 (2022-10)
Expediting Large-Scale Vision Transformer for Dense Prediction without Fine-tuning
2310.15171 (2023-10)
RoboDepth: Robust Out-of-Distribution Depth Estimation under Corruptions
and 1 more not named here
attn_cosine_sim 3 other papers
2 dated earlier, 1 later
2201.00424 (2022-01)
Splicing ViT Features for Semantic Appearance Transfer
2204.00928 (2022-04)
SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image
2401.09146 (2024-01)
Continuous Piecewise-Affine Based Motion Model for Image Animation
init_grid_param 2 other papers
2 later
2309.15426 (2023-09)
NeuRBF: A Neural Fields Representation with Adaptive Radial Basis Functions
2404.02742 (2024-04)
LiDAR4D: Dynamic Neural Fields for Novel Space-time View LiDAR Synthesis
DiffAugment 1 other paper
1 dated earlier
2204.00928 (2022-04)
SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image
get_normalized_directions 1 other paper
1 later
2309.15426 (2023-09)
NeuRBF: A Neural Fields Representation with Adaptive Radial Basis Functions
normalize_aabb 1 other paper
1 later
2309.15426 (2023-09)
NeuRBF: A Neural Fields Representation with Adaptive Radial Basis Functions
reparametrize 1 other paper
1 dated earlier
2204.00928 (2022-04)
SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Image

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 radiance field (NeRF) shows powerful performance in novel view synthesis and 3D geometry reconstruction, but it suffers from critical performance degradation when the number of known viewpoints is drastically reduced. Existing works attempt to overcome this problem by employing external priors, but their success is limited to certain types of scenes or datasets. Employing monocular depth estimation (MDE) networks, pretrained on large-scale RGB-D datasets, with powerful generalization capability would be a key to solving this problem: however, using MDE in conjunction with NeRF comes with a new set of challenges due to various ambiguity problems exhibited by monocular depths. In this light, we propose a novel framework, dubbed DäRF, that achieves robust NeRF reconstruction with a handful of real-world images by combining the strengths of NeRF and monocular depth estimation through online complementary training. Our framework imposes the MDE network's powerful geometry prior to NeRF representation at both seen and unseen viewpoints to enhance its robustness and coherence. In addition, we overcome the ambiguity problems of monocular depths through patch-wise scaleshift fitting and geometry distillation, which adapts the MDE network to produce depths aligned accurately with NeRF geometry. Experiments show our framework achieves state-of-the-art results both quantitatively and qualitatively, demonstrating consistent and reliable performance in both indoor and outdoor real-world datasets. Project page is available at https://ku-cvlab.github.io/DaRF/.

For agents

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

get_harvested_code_for_paper("2305.19201")
get_code_for_paper("2305.19201")
have("2305.19201")

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

curl https://syntology.ai/api/ran/2305.19201.json

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

[![Syntology run record](https://syntology.ai/api/ran/2305.19201.svg)](https://syntology.ai/paper/2305.19201)

Connect an agent — have() is free.