SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.20459 · 2026

PRESCIENCE: A Dataset and Benchmark for Scientific Forecasting

Oyvind Tafjord, Amanpreet Singh, James Evans, Austin Kozlowski, Doug Downey, Tom Hope, Daniel Weld, Anirudh Ajith, Jay Deyoung, Nadav Kunievsky, Lacer Judge Prompt

arXiv · PDF · Open in the Atlas

Code that ran

22 of 29 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; 5 honoured the contract we drafted, 7 violated it and 0 have no contract verdict; 10 failures of our instrument; 7 no recorded run), measured 2026-11-22; 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/2602.20459.json · badge: /api/ran/2602.20459.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
RomGai/BrainVis canonical (classifier label)
classifier basis: census
8 ran · 4 no recorded run
LHDuan/ConDaFormer canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
HKUDS/LightRAG canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
FunctionStatusWhere it lives
batch2offset Ran
ran; our contract draft was wrong, not the code
LHDuan/ConDaFormer/pcr/models/utils.py
pointer only (licence: NONE) · get_code("2a2c9e53188de780")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

channel_last Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
RomGai/BrainVis/cascade_diffusion.py
code served (permissive licence) · get_code("979a7927f4aa11d9")
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 (4, 5, 3) 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 einops import rearrange

torch.manual_seed(0)

# The function channel_last takes an image tensor and checks if it's already
# in channel-last format (channels in last dimension). If not, it rearranges
# from 'c h w' format to 'h w c' format.

# Test case 1: Image already in channel-last format (h, w, c)
img_channel_last = torch.randn(4, 5, 3)

# Test case 2: Image in channel-first format (c, h, w)
img_channel_first = torch.randn(3, 4, 5)

# We'll use the channel-first case as the primary test
ARGS = {
    'img': img_channel_first
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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 (4, 5, 3) — shape (4, 5, 3) 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
collect_kg_merge_candidates Ran
ran; our contract draft was wrong, not the code
HKUDS/LightRAG/lightrag/operate.py
code served (permissive licence) · get_code("014de214db1895ca")
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 torch

# Construct a minimal chunk_results list
# Each element is a tuple of (maybe_nodes, maybe_edges)
# maybe_nodes is a dict with entity names as keys
# maybe_edges is a dict with edge keys as tuples (source, target)

chunk_results = [
    (
        {
            "entity_alice": {"type": "person"},
            "entity_bob": {"type": "person"},
        },
        {
            ("entity_alice", "entity_bob"): {"relation": "knows"},
            ("entity_bob", "entity_alice"): {"relation": "knows"},
        },
    ),
    (
        {
            "entity_charlie": {"type": "person"},
            "entity_alice": {"type": "person"},
        },
        {
            ("entity_charlie", "entity_alice"): {"relation": "works_with"},
        },
    ),
]

ARGS = {
    "chunk_results": chunk_results,
}
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 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
find_target_span Ran
ran, honoured the contract we drafted
HKUDS/LightRAG/lightrag/multimodal_context.py
code served (permissive licence) · get_code("1729c5aef344cc43")
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 re

# Test case 1: drawing with simple ID
kind_1 = "drawings"
item_id_1 = "fig_001"
block_content_1 = '<drawing id="fig_001" src="image.png"/>'

# Test case 2: table with complex formatting
kind_2 = "tables"
item_id_2 = "tbl_42"
block_content_2 = '<table id="tbl_42" class="data-table"><tr><td>Cell</td></tr></table>'

# Test case 3: equation with nested content
kind_3 = "equations"
item_id_3 = "eq_alpha"
block_content_3 = '<equation id="eq_alpha" type="display">E=mc^2</equation>'

# Test case 4: ID with special regex characters that need escaping
kind_4 = "drawings"
item_id_4 = "fig_[special]"
block_content_4 = '<drawing id="fig_[special]" data-src="test.svg"/>'

# Test case 5: Not found case
kind_5 = "tables"
item_id_5 = "missing_id"
block_content_5 = '<table id="other_id"><tr><td>Data</td></tr></table>'

# Test case 6: Invalid kind (should return None)
kind_6 = "invalid_kind"
item_id_6 = "any_id"
block_content_6 = '<drawing id="any_id"/>'

ARGS = {
    "kind": "drawings",
    "item_id": "fig_001",
    "block_content": '<drawing id="fig_001" src="image.png"/>'
}
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 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
is_namespace Ran
ran, violated the contract we drafted
HKUDS/LightRAG/lightrag/namespace.py
code served (permissive licence) · get_code("cad2192dedaa609f")
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 torch
import numpy as np
from typing import Iterable

def is_namespace(namespace: str, base_namespace: str | Iterable[str]):
    if isinstance(base_namespace, str):
        return namespace.endswith(base_namespace)
    return any((is_namespace(namespace, ns) for ns in base_namespace))

# Test case 1: namespace is a string, base_namespace is a string
namespace_1 = "model.layer.weight"
base_namespace_1 = "weight"

# Test case 2: namespace is a string, base_namespace is an iterable of strings
namespace_2 = "model.layer.bias"
base_namespace_2 = ["weight", "bias", "norm"]

ARGS = {
    "namespace": namespace_1,
    "base_namespace": base_namespace_1
}
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 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
isnan Ran
ran, violated the contract we drafted
LHDuan/ConDaFormer/pcr/utils/losses.py
pointer only (licence: NONE); the stored copy get_code serves, from sidak/otfusion, is pointer only (licence: NONE) · get_code("5f4f4355be28c01e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

larger_the_better Ran
ran, violated the contract we drafted
RomGai/BrainVis/eval_metrics.py
code served (permissive licence) · get_code("3fb22d63b664978c")
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 numpy as np

torch.manual_seed(0)

# The function simply compares two values using the > operator
# Both gt and comp should be comparable (numbers, tensors, arrays, etc.)

# Create small tensors for comparison
gt = torch.tensor([[1.0, 2.0, 3.0, 4.0],
                    [5.0, 6.0, 7.0, 8.0]])

comp = torch.tensor([[0.5, 1.5, 2.5, 3.5],
                      [4.5, 5.5, 6.5, 7.5]])

ARGS = {
    "gt": gt,
    "comp": comp
}
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 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
load_content_rows_by_blockid Ran
ran; our contract draft was wrong, not the code
HKUDS/LightRAG/lightrag/multimodal_context.py
code served (permissive licence) · get_code("c629108e63d2245a")
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 json
import tempfile
from pathlib import Path

# Create a temporary blocks.jsonl file with test data
temp_dir = tempfile.mkdtemp()
blocks_path = Path(temp_dir) / "blocks.jsonl"

# Write test JSONL data
test_data = [
    {"type": "content", "blockid": "block1", "content": "This is content for block 1"},
    {"type": "content", "blockid": "block2", "content": "This is content for block 2"},
    {"type": "other", "blockid": "block3", "content": "This should be ignored"},
    {"type": "content", "blockid": "block1", "content": "Duplicate block1, should be ignored"},
    {"type": "content", "blockid": "block4", "content": "Content for block 4"},
    {"type": "content", "blockid": "", "content": "Empty blockid, should be ignored"},
    {"type": "content", "blockid": "block5"},  # Missing content
    "not a dict",  # Invalid JSON object
    '{"type": "content", "blockid": "block6", "content": "Block 6 content"}',
]

with blocks_path.open('w', encoding='utf-8') as f:
    for item in test_data:
        if isinstance(item, dict):
            f.write(json.dumps(item) + '\n')
        else:
            f.write(item + '\n')

ARGS = {
    "blocks_path": str(blocks_path)
}
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
lovasz_grad Ran
ran, violated the contract we drafted
LHDuan/ConDaFormer/pcr/utils/losses.py
pointer only (licence: NONE) · get_code("e85b5d21dc5b8382")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

mse_metric Ran
ran, honoured the contract we drafted
RomGai/BrainVis/eval_metrics.py
code served (permissive licence) · get_code("26ee4a53b6d00e3d")
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 float64, ndim 0, 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 numpy as np

# Create two small image arrays
torch_seed = 0  # For reproducibility if needed

img1 = np.random.randn(2, 4, 3, 3)  # batch=2, height=4, width=3, channels=3
img2 = np.random.randn(2, 4, 3, 3)  # same shape as img1

ARGS = {
    "img1": img1,
    "img2": img2,
}
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 float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
normalize Ran
ran, violated the contract we drafted
RomGai/BrainVis/cascade_diffusion.py
code served (permissive licence) · get_code("256f9a7da1ea965e")
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 (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
from einops import rearrange, repeat

torch.manual_seed(0)

# The function takes a single parameter 'img'
# Looking at the function body:
# - img.shape[-1] is checked (needs to be indexable)
# - rearrange is called if shape[-1] == 3
# - torch.tensor() is called on img
# - img is multiplied by 2.0 and subtracted 1.0
# So img should be array-like with numeric values

# Create a small image with 3 channels (H=4, W=5, C=3)
img = np.random.rand(4, 5, 3).astype(np.float32)

ARGS = {
    "img": img
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, 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 3, dtype family float, shape (3, 4, 5) — shape (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
normalize_array Ran
ran, violated the contract we drafted
RomGai/BrainVis/datautils.py
code served (permissive licence) · get_code("e7cb1387b8010919")
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 ndarray, ndim 1, dtype family float, shape (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 numpy as np

# Create a small numpy array with distinct values for testing
arr = np.array([1.0, 2.0, 3.0, 4.0, 5.0])

ARGS = {
    "arr": arr
}
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 ndarray, ndim 1, dtype family float, shape (5,) — shape (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
normalize_chunk_heading Ran
ran; our contract draft was wrong, not the code
HKUDS/LightRAG/lightrag/chunk_schema.py
code served (permissive licence) · get_code("245793d97f7a353f")
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)

# Test case 1: nested dict format with all fields
dp_nested_full = {
    'heading': {
        'level': 2,
        'heading': 'Introduction',
        'parent_headings': ['Chapter 1', 'Section A']
    }
}

# Test case 2: legacy flat format
dp_legacy_flat = {
    'heading': 'Methods',
    'level': 1,
    'parent_headings': ['Background']
}

# Test case 3: empty/missing inputs should return None
dp_empty = {
    'heading': '',
    'level': 0,
    'parent_headings': []
}

# Test case 4: nested dict with some empty fields
dp_nested_partial = {
    'heading': {
        'level': 3,
        'heading': 'Subsection',
        'parent_headings': None
    }
}

# Test case 5: legacy format with missing parent_headings
dp_legacy_minimal = {
    'heading': 'Results',
    'level': 2
}

# Use one of the test cases as the main argument
ARGS = {
    'dp': dp_nested_full
}
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
normalize_kv_create_time Ran
ran, honoured the contract we drafted
HKUDS/LightRAG/lightrag/base.py
code served (permissive licence) · get_code("65781232be7b3546")
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 logging
from typing import Any

# Setup logger since the function uses it
logger = logging.getLogger(__name__)
logger.setLevel(logging.WARNING)

# Test cases covering the main branches:
# 1. None input
# 2. Boolean input (special case)
# 3. Valid int
# 4. Valid float that can be coerced to int
# 5. String that can be coerced to int
# 6. Invalid type that cannot be coerced

ARGS = {
    "value": 1609459200
}
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
offset2batch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LHDuan/ConDaFormer/pcr/models/condaformer/condaformer.py
pointer only (licence: NONE) · get_code("6dd8eac8666461f5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

offset2batch Ran
ran, honoured the contract we drafted
LHDuan/ConDaFormer/pcr/models/utils.py
pointer only (licence: NONE) · get_code("d78aee4794314c59")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

preprocess_images Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
RomGai/BrainVis/evaluate_GA.py
code served (permissive licence) · get_code("704d18c7ee43b72e")
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, 3) 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 PIL import Image
import numpy as np

torch.manual_seed(0)

# Create small dummy images as PIL Images
# The function expects a list of images that can be passed to ToTensor()
images = []
for i in range(2):
    # Create a small 4x3 RGB image
    img_array = np.random.randint(0, 256, (4, 3, 3), dtype=np.uint8)
    img = Image.fromarray(img_array, mode='RGB')
    images.append(img)

ARGS = {
    'images': images
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy, torch, torchvision — 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, 3) — shape (2, 3, 4, 3) 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
recalls_and_ndcgs_for_ks Ran
ran; our contract draft was wrong, not the code
RomGai/BrainVis/loss.py
code served (permissive licence) · get_code("c59c3677a8ada751")
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

torch.manual_seed(0)

# scores: tensor of shape (batch_size, num_items) with similarity/relevance scores
scores = torch.randn(2, 4)

# answers: tensor of shape (batch_size,) or list of lists containing indices of relevant items
answers = torch.tensor([[0, 2], [1, 3]])

# k: number of top items to consider for recall/ndcg computation
k = 3

ARGS = {
    "scores": scores,
    "answers": answers,
    "k": k,
}
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
smaller_the_better Ran
ran, violated the contract we drafted
RomGai/BrainVis/eval_metrics.py
code served (permissive licence) · get_code("2e211cdb44f1535e")
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 torch
import numpy as np

torch.manual_seed(0)

# The function smaller_the_better takes two arguments: gt and comp
# It performs a less-than comparison: gt < comp
# This works with any comparable objects (numbers, arrays, tensors, etc.)

# Create simple scalar values
gt_value = 5
comp_value = 10

ARGS = {
    "gt": gt_value,
    "comp": comp_value
}
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 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
tmp_path_for Ran
ran; our contract draft was wrong, not the code
HKUDS/LightRAG/lightrag/file_atomic.py
code served (permissive licence) · get_code("851727a852a7d324")
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 os
import threading
import time

ARGS = {
    "file_name": "model_checkpoint.pt"
}
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
to_3d_numpy Ran
ran, honoured the contract we drafted
LHDuan/ConDaFormer/pcr/models/condaformer/utils.py
pointer only (licence: NONE) · get_code("ae72b160c77cb3e4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

trunc_normal_ Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
LHDuan/ConDaFormer/pcr/models/utils.py
pointer only (licence: NONE) · get_code("5436174f8c0e64e0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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.

fit_lr Not yet run
unverified (harvested, no recorded run)
RomGai/BrainVis/classification.py
code served (permissive licence) · get_code("d354e00aa94a14de")
How it ran

no invocation recorded.

flush_may_have_lost_reference Not yet run
unverified (harvested, no recorded run)
HKUDS/LightRAG/lightrag/exceptions.py
code served (permissive licence) · get_code("07fbfec7874e6a96")
How it ran

no invocation recorded.

grid_sample Not yet run
unverified (harvested, no recorded run)
LHDuan/ConDaFormer/pcr/models/condaformer/utils.py
pointer only (licence: NONE) · get_code("66d6bb150bbb2e57")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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. No invocation recorded.

load_EEG Not yet run
unverified (harvested, no recorded run)
RomGai/BrainVis/datautils.py
code served (permissive licence) · get_code("8ed8f44f5b0d2921")
How it ran

no invocation recorded.

load_EEG_with_img_name Not yet run
unverified (harvested, no recorded run)
RomGai/BrainVis/datautils.py
code served (permissive licence) · get_code("bbe6b674d1af4f5b")
How it ran

no invocation recorded.

lovasz_softmax Not yet run
unverified (harvested, no recorded run)
LHDuan/ConDaFormer/pcr/utils/losses.py
pointer only (licence: NONE) · get_code("82b9b0ff4b3626c9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, LHDuan/ConDaFormer 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. No invocation recorded.

ssim_metric Not yet run
unverified (harvested, no recorded run)
RomGai/BrainVis/evaluate_GA.py
code served (permissive licence) · get_code("ccd0f27cd6e9b5ba")
How it ran

no invocation recorded.

Also present in other papers' code

Of 29 samples lifted from this paper's linked repositories, 29 carry a function identity and 0 do not. Of the 29 distinct functions with an identity, 21 are also present in other papers whose code we hold (21 with at least one paper dated earlier, 2 the same month, 1 later) and 8 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
trunc_normal_ 160 other papers
154 dated earlier, 3 the same month, 3 later · 36 only as an assembly brick
1911.10194 (2019-11)
Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation
2005.10242 (2020-05)
Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
2007.12911 (2020-07)
Tighter risk certificates for neural networks
and 157 more not named here
lovasz_grad 40 other papers
38 dated earlier, 2 the same month · 1 only as an assembly brick
1505.04597 (2015-05)
U-Net: Convolutional Networks for Biomedical Image Segmentation
1807.10165 (2018-07)
UNet++: A Nested U-Net Architecture for Medical Image Segmentation
1811.11721 (2018-11)
CCNet: Criss-Cross Attention for Semantic Segmentation
and 37 more not named here
isnan 16 other papers
16 dated earlier · 2 only as an assembly brick
1505.04597 (2015-05)
U-Net: Convolutional Networks for Biomedical Image Segmentation
1910.05653 (2019-10)
Model Fusion via Optimal Transport
2008.07043 (2020-08)
Oriented Object Detection in Aerial Images with Box Boundary-Aware Vectors
and 13 more not named here
batch2offset 2 other papers
2 dated earlier
2312.11112 (2023-12)
ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding
2510.09659 (2025-10)
Heterogeneous Point Set Transformers for Segmentation of Multiple View Particle Detectors
is_namespace 2 other papers
2 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
2025.findings-emnlp.290 (2025 (year only))
ReGraphRAG: Reorganizing Fragmented Knowledge Graphs for Multi-Perspective Retrieval-Augmented Generation
offset2batch 2 other papers
2 dated earlier
2307.09316 (2023-07)
MarS3D: A Plug-and-Play Motion-Aware Model for Semantic Segmentation on Multi-Scan 3D Point Clouds
2312.11112 (2023-12)
ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding
channel_last 1 other paper
1 dated earlier
2211.06956 (2022-11)
Seeing Beyond the Brain: Conditional Diffusion Model with Sparse Masked Modeling for Vision Decoding
collect_kg_merge_candidates 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
find_target_span 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
fit_lr 1 other paper
1 dated earlier
2512.00239 (2025-12)
Self-Supervised Dynamical System Representations for Physiological Time-Series
flush_may_have_lost_reference 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
grid_sample 1 other paper
1 dated earlier
2312.11112 (2023-12)
ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding
load_content_rows_by_blockid 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
lovasz_softmax 1 other paper
1 dated earlier
2312.11112 (2023-12)
ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding
normalize 1 other paper
1 dated earlier
2211.06956 (2022-11)
Seeing Beyond the Brain: Conditional Diffusion Model with Sparse Masked Modeling for Vision Decoding
normalize_chunk_heading 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
normalize_kv_create_time 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation
offset2batch 1 other paper
1 dated earlier
2312.11112 (2023-12)
ConDaFormer: Disassembled Transformer with Local Structure Enhancement for 3D Point Cloud Understanding
recalls_and_ndcgs_for_ks 1 other paper
1 dated earlier
2512.00239 (2025-12)
Self-Supervised Dynamical System Representations for Physiological Time-Series
tmp_path_for 1 other paper
1 dated earlier
2410.05779 (2024-10)
LightRAG: Simple and Fast Retrieval-Augmented Generation

Shown: the 20 functions present in the most other papers. 1 more shared functions are not listed here.

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

Can AI systems trained on the existing scientific record forecast the advances that will follow? We introduce PRESCIENCE, a dataset and benchmark for scientific forecasting built around 98K recent AI research papers, together with companion papers covering author publication histories and citation links, yielding 502K papers in total. The resulting paper records include titles, abstracts, disambiguated author identities, influential references, topic labels, citation trajectories, and metadata snapshotted to respect temporal cutoffs. We instantiate seven exemplar tasks: five paper-anchored tasks-contribution generation, collaborator prediction, prior work selection, citation count prediction, and future combination prediction-and two aggregate topic trend forecasting variants. We develop baselines ranging from simple heuristics and embedding methods to frontier language models and agentic systems, and introduce LACER, an LLM-based metric for evaluating similarity of generated contribution descriptions that agrees better with human judgments than existing metrics. Finally, we compose task models to generate a 12-month synthetic corpus and find that the resulting papers are systematically less diverse and less novel than human-authored research from the same period.

For agents

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

get_harvested_code_for_paper("2602.20459")
get_code_for_paper("2602.20459")
have("2602.20459")

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

curl https://syntology.ai/api/ran/2602.20459.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.20459.svg)](https://syntology.ai/paper/2602.20459)

Connect an agent — have() is free.