SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2111.05458 · 2021

Which priors matter? Benchmarking models for learning latent dynamics

arXiv · PDF · Open in the Atlas

Code that ran

7 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 2 failures of our instrument; 4 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/2111.05458.json · badge: /api/ran/2111.05458.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
westny/neural-stability community (classifier label)
classifier basis not recorded on the link
7 ran · 4 no recorded run
FunctionStatusWhere it lives
create_hdf5_dataset Ran
ran on a synthesized input
westny/neural-stability/latent_dynamics/tf_converter_hdf5.py
code served (permissive licence) · get_code("4a9ef9984aad0d97")
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 Dataset. 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 h5py
import tempfile
import os

# Create a temporary HDF5 file for testing
temp_dir = tempfile.mkdtemp()
temp_file_path = os.path.join(temp_dir, "test.h5")
file = h5py.File(temp_file_path, "w")

ARGS = {
    "file": file,
    "name": "dataset_name",
    "shape": (4, 3),
    "dtype": "float32"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
h5py — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Dataset — a Dataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Dataset output the recorded ndim 2 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
import_from_module Ran
ran; our contract draft was wrong, not the code
westny/neural-stability/preamble.py
code served (permissive licence) · get_code("c7ed721f4fc7e643")
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 type. 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 importlib
from typing import Callable

torch = None
try:
    import torch
except ImportError:
    pass

# The function import_from_module takes two string parameters:
# 1. module_name: str - the name of the module to import from
# 2. class_name: str - the name of the class/callable to retrieve from that module

# Since the function uses importlib.import_module and getattr to dynamically
# import and retrieve a class/callable, we need to provide valid module and class names
# that actually exist in Python's standard library or commonly available packages.

# Let's use torch.nn as our module and Linear as our class, since torch should be available
ARGS = {
    "module_name": "torch.nn",
    "class_name": "Linear"
}
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 type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
import_module Ran
ran; our contract draft was wrong, not the code
westny/neural-stability/preamble.py
code served (permissive licence) · get_code("cc34274b1d241296")
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 module. 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 importlib

torch_module = importlib.import_module('torch')

ARGS = {
    'module_name': 'math'
}
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 module — a module output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a module 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
load_config Ran
ran on a synthesized input
westny/neural-stability/preamble.py
code served (permissive licence) · get_code("92c60eff70817c99")
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 directory structure with a config file
temp_dir = tempfile.TemporaryDirectory()
configs_path = Path(temp_dir.name) / 'configs'
configs_path.mkdir()

# Create a subdirectory
subdir = configs_path / 'models'
subdir.mkdir()

# Create a sample config file
config_file = subdir / 'test_config.json'
sample_config = {
    'model': 'resnet50',
    'learning_rate': 0.001,
    'batch_size': 32,
    'epochs': 10
}
with open(config_file, 'w', encoding='utf-8') as f:
    json.dump(sample_config, f)

# Change to temp directory so relative path works
import os
original_cwd = os.getcwd()
os.chdir(temp_dir.name)

ARGS = {
    'config': 'test_config'
}
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
ortho_matrix Ran
ran on a synthesized input
westny/neural-stability/stb_init.py
code served (permissive licence) · get_code("2ab8be501247153a")
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 (5, 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 math

torch.manual_seed(0)

ARGS = {
    "n": 5,
    "use_imag": False,
}
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 (5, 5) — shape (5, 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
sample_poles Ran
ran on a synthesized input
westny/neural-stability/stb_init.py
code served (permissive licence) · get_code("98bac5814c1eef56")
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 (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
import math

torch.manual_seed(0)
np.random.seed(0)

ARGS = {
    "n_states": 5,
    "step_size": 0.1,
    "solver_order": 2,
    "use_imag": True,
    "eps": -0.1,
    "exclusion_order": 0,
}
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 (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
str_to_bool Ran
ran on a synthesized input
westny/neural-stability/arguments.py
code served (permissive licence) · get_code("71d07db29119f009")
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
from argparse import ArgumentTypeError

def str_to_bool(value: bool | str) -> bool:
    """Used for boolean arguments in argparse; avoiding `store_true` and `store_false`."""
    true_vals = ('yes', 'true', 't', 'y', '1')
    false_vals = ('no', 'false', 'f', 'n', '0')
    if isinstance(value, bool):
        return value
    if value.lower() in true_vals:
        return True
    if value.lower() in false_vals:
        return False
    raise ArgumentTypeError('Boolean value expected.')


ARGS = {
    'value': 'true'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
collate_fn_img Not yet run
unverified (harvested, no recorded run)
westny/neural-stability/latent_dynamics/data_utils.py
code served (permissive licence) · get_code("eb6680dca1c08000")
How it ran

no invocation recorded.

collate_fn_state Not yet run
unverified (harvested, no recorded run)
westny/neural-stability/latent_dynamics/data_utils.py
code served (permissive licence) · get_code("f2bc3a70a36b7d38")
How it ran

no invocation recorded.

sample_exclusive_poles Not yet run
unverified (harvested, no recorded run)
westny/neural-stability/stb_init.py
code served (permissive licence) · get_code("7cdf1377f0fc42ca")
How it ran

no invocation recorded.

switch Not yet run
unverified (harvested, no recorded run)
westny/neural-stability/latent_dynamics/data_utils.py
code served (permissive licence) · get_code("023376a296c28fe0")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 0 later) and 9 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
import_from_module 1 other paper
1 dated earlier
1911.07602 (2019-11)
The inD Dataset: A Drone Dataset of Naturalistic Road User Trajectories at German Intersections
import_module 1 other paper
1 dated earlier
1911.07602 (2019-11)
The inD Dataset: A Drone Dataset of Naturalistic Road User Trajectories at German Intersections

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

Learning dynamics is at the heart of many important applications of machine learning (ML), such as robotics and autonomous driving. In these settings, ML algorithms typically need to reason about a physical system using high dimensional observations, such as images, without access to the underlying state. Recently, several methods have proposed to integrate priors from classical mechanics into ML models to address the challenge of physical reasoning from images. In this work, we take a sober look at the current capabilities of these models. To this end, we introduce a suite consisting of 17 datasets with visual observations based on physical systems exhibiting a wide range of dynamics. We conduct a thorough and detailed comparison of the major classes of physically inspired methods alongside several strong baselines. While models that incorporate physical priors can often learn latent spaces with desirable properties, our results demonstrate that these methods fail to significantly improve upon standard techniques. Nonetheless, we find that the use of continuous and time-reversible dynamics benefits models of all classes.

For agents

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

get_harvested_code_for_paper("2111.05458")
get_code_for_paper("2111.05458")
have("2111.05458")

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

curl https://syntology.ai/api/ran/2111.05458.json

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

[![Syntology run record](https://syntology.ai/api/ran/2111.05458.svg)](https://syntology.ai/paper/2111.05458)

Connect an agent — have() is free.