SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.09100 · NeurIPS · 2024

Retrospective for the Dynamic Sensorium Competition for predicting large-scale mouse primary visual cortex activity from videos

Lei Ma, Tiejun Huang, Fabian Sinz, Andreas Tolias, Alexander Ecker, Konstantin Willeke, Yu Zhu, Paul Fahey, Mohammad Bashiri, Zhiwei Ding, Kayla Ponder, Bryan Li, and 13 more

arXiv · PDF · Open in the Atlas

Code that ran

17 of 21 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 16 failures of our instrument; 4 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/2407.09100.json · badge: /api/ran/2407.09100.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
bryanlimy/ViV1T canonical (classifier label)
classifier basis: census
7 ran · 2 no recorded run
zhuyu-cs/Dynamic-VFM canonical (classifier label)
classifier basis: census
6 ran · 2 no recorded run
ecker-lab/sensorium_2023 canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
lRomul/sensorium canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
check_hyperparam_for_layers Ran
ran, honoured the contract we drafted
zhuyu-cs/Dynamic-VFM/models/conv3d_core.py
pointer only (licence: NONE) · get_code("f6b87b79a83d122c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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.

check_output Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/utils/utils.py
code served (permissive licence) · get_code("cf761ce0ca742b66")
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 subprocess

def check_output(command: list):
    """Run command in subprocess and return output as string"""
    return subprocess.check_output(command).strip().decode()

# Construct valid arguments for check_output
# The function expects a list representing a shell command
# Using a simple echo command as a test case
ARGS = {
    "command": ["echo", "hello world"]
}
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
correlation Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/metrics.py
code served (permissive licence) · get_code("3bbc3729e03c6d6c")
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 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create small torch tensors for testing
y1_torch = torch.randn(2, 4)
y2_torch = torch.randn(2, 4)

# Create small numpy arrays for testing
y1_numpy = np.random.randn(2, 4)
y2_numpy = np.random.randn(2, 4)

# Test with torch tensors
ARGS = {
    "y1": y1_torch,
    "y2": y2_torch,
    "dim": -1,
    "eps": 1e-08,
}
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 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
drop_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lRomul/sensorium/src/models/dwiseneuro.py
code served (permissive licence) · get_code("971ae8d2d8313e30")
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) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 3, 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

torch.manual_seed(0)

# Create a small tensor for testing
# Shape: (batch_size=2, channels=4, height=3, width=3) - 4D tensor
x = torch.randn(2, 4, 3, 3)

ARGS = {
    "x": x,
    "drop_prob": 0.5,
    "training": True,
    "scale_by_keep": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 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
get_device Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/utils/utils.py
code served (permissive licence) · get_code("b98b672b95d88ebe")
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 device. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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

torch.manual_seed(0)

# The function takes an optional device string parameter
# It returns a torch.device object
# The parameter can be None, 'cpu', 'cuda', or 'mps'

ARGS = {
    'device': None
}
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 device — a device output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a device 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
get_dims_for_loader_dict Ran
ran; our contract draft was wrong, not the code
ecker-lab/sensorium_2023/sensorium/models/utility.py
pointer only (licence: NONE) · get_code("aaf130cf62109aaa")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ecker-lab/sensorium_2023 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_dims_for_loader_dict Ran
ran; our contract draft was wrong, not the code
zhuyu-cs/Dynamic-VFM/models/utility.py
pointer only (licence: NONE) · get_code("325390bf74ae81ee")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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_io_dims Ran
ran; our contract draft was wrong, not the code
ecker-lab/sensorium_2023/sensorium/models/utility.py
pointer only (licence: NONE) · get_code("4d1711af341a89b8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ecker-lab/sensorium_2023 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_io_dims Ran
ran; our contract draft was wrong, not the code
zhuyu-cs/Dynamic-VFM/models/utility.py
pointer only (licence: NONE) · get_code("d3f6c644aed06d3f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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_module_output Ran
ran; our contract draft was wrong, not the code
zhuyu-cs/Dynamic-VFM/models/make_model_adaptive.py
pointer only (licence: NONE) · get_code("7486444a4b765a39")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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_output_shape Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/model/critic.py
code served (permissive licence) · get_code("2387de91debc2c22")
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

ARGS = {
    'input_shape': (2, 16, 20, 24),
    'out_channels': 8,
    'kernel_size': 3,
    'stride': 1,
    'padding': 0,
    'dilation': None,
}
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, 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
poisson_loss Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/metrics.py
code served (permissive licence) · get_code("8185ec50f5bd1985")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create small tensors for y_true and y_pred
# Shape: (batch=2, frames=3, neurons=4)
y_true = torch.rand(2, 3, 4) + 0.1  # Add offset to avoid zeros
y_pred = torch.rand(2, 3, 4) + 0.1

# eps can be a float or tensor
eps = 1e-08

# reduction must be 'sum' or 'mean'
reduction = 'sum'

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
    'eps': eps,
    'reduction': reduction
}
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
prepare_grid Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ecker-lab/sensorium_2023/sensorium/models/utility.py
pointer only (licence: NONE) · get_code("4a3eb40ef4963538")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ecker-lab/sensorium_2023 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.

prepare_grid Ran
ran; our contract draft was wrong, not the code
zhuyu-cs/Dynamic-VFM/models/utility.py
pointer only (licence: NONE) · get_code("3dfa8cee58b9b8ad")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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.

register Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/criterions.py
code served (permissive licence) · get_code("3c702d811d125a17")
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 function. 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)

# The register function takes one parameter: name (a string)
# It returns a decorator function that adds functions to the _CRITERION dict
# To test this, we need to call register with a name argument

name = "loss_function"

ARGS = {
    "name": name
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
shifted_window_attention_3d Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhuyu-cs/Dynamic-VFM/models/swin_core.py
pointer only (licence: NONE) · get_code("4f1b2ed2a3391b71")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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.

sse Ran
ran; our contract draft was wrong, not the code
bryanlimy/ViV1T/src/viv1t/metrics.py
code served (permissive licence) · get_code("657ce70f94e81f45")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Create small tensors for y_true and y_pred
# Shape: (batch=2, frames=3, neurons=4)
y_true = torch.randn(2, 3, 4)
y_pred = torch.randn(2, 3, 4)

# reduction parameter can be 'mean' or 'sum'
reduction = 'mean'

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
    'reduction': reduction,
}
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_model_info Not yet run
unverified (harvested, no recorded run)
bryanlimy/ViV1T/src/viv1t/model/model.py
code served (permissive licence) · get_code("9448aaf59e5342b1")
How it ran

no invocation recorded.

make_video_model_adaptive Not yet run
unverified (harvested, no recorded run)
zhuyu-cs/Dynamic-VFM/models/make_model_adaptive.py
pointer only (licence: NONE) · get_code("8b8ce82c52c4e78c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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.

make_video_model_mean Not yet run
unverified (harvested, no recorded run)
zhuyu-cs/Dynamic-VFM/models/make_model_mean.py
pointer only (licence: NONE) · get_code("a244a5150de05c50")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhuyu-cs/Dynamic-VFM 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.

support_bf16 Not yet run
unverified (harvested, no recorded run)
bryanlimy/ViV1T/src/viv1t/utils/utils.py
code served (permissive licence) · get_code("e07554b34923da1b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 1 later) and 19 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
drop_path 66 other papers
30 dated earlier, 4 the same month, 29 later, 1 same year, order not determinable, 2 undated · 15 only as an assembly brick
1906.01502 (2019-06)
How multilingual is Multilingual BERT?
2011.10566 (2020-11)
Exploring Simple Siamese Representation Learning
2103.03230 (2021-03)
Barlow Twins: Self-Supervised Learning via Redundancy Reduction
and 63 more not named here
register 1 other paper
1 dated earlier
2302.03023 (2023-02)
V1T: large-scale mouse V1 response prediction using a Vision Transformer

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

Understanding how biological visual systems process information is challenging because of the nonlinear relationship between visual input and neuronal responses. Artificial neural networks allow computational neuroscientists to create predictive models that connect biological and machine vision. Machine learning has benefited tremendously from benchmarks that compare different model on the same task under standardized conditions. However, there was no standardized benchmark to identify state-of-the-art dynamic models of the mouse visual system. To address this gap, we established the SENSORIUM 2023 Benchmark Competition with dynamic input, featuring a new large-scale dataset from the primary visual cortex of ten mice. This dataset includes responses from 78,853 neurons to 2 hours of dynamic stimuli per neuron, together with the behavioral measurements such as running speed, pupil dilation, and eye movements. The competition ranked models in two tracks based on predictive performance for neuronal responses on a held-out test set: one focusing on predicting in-domain natural stimuli and another on out-of-distribution (OOD) stimuli to assess model generalization. As part of the NeurIPS 2023 competition track, we received more than 160 model submissions from 22 teams. Several new architectures for predictive models were proposed, and the winning teams improved the previous state-of-the-art model by 50%. Access to the dataset as well as the benchmarking infrastructure will remain online at www.sensorium-competition.net. Preprint. Under review.

For agents

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

get_harvested_code_for_paper("2407.09100")
get_code_for_paper("2407.09100")
have("2407.09100")

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

curl https://syntology.ai/api/ran/2407.09100.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.09100.svg)](https://syntology.ai/paper/2407.09100)

Connect an agent — have() is free.