SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.15812 · ICLR · 2025

UNCOVERING NONLINEAR CONNECTIONS IN MULTIMODAL REPRESENTATIONS

Luca Bortolussi, Park, Lorenzo Basile, Santiago Acevedo, Fabio Anselmi, Alex Rodriguez

arXiv · PDF · Open in the Atlas

Code that ran

9 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 6 failures of our instrument; 2 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/2406.15812.json · badge: /api/ran/2406.15812.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
lorenzobasile/IDCorrelation marked official in Papers with Code
Papers with Code archive
6 ran · 1 no recorded run
brando90/ultimate-anatome extension (classifier label)
classifier basis: classifier_2025
3 ran · 1 no recorded run
FunctionStatusWhere it lives
MLE Ran
ran; our contract draft was wrong, not the code
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("052b8393843a4c83")
How it ran

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

add_fourier_noise Ran
ran; our contract draft was wrong, not the code
brando90/ultimate-anatome/anatome/fourier.py
code served (permissive licence) · get_code("a612604b8ff10701")
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, 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
from torch import Tensor
import torch.nn.functional as F

torch.manual_seed(0)

# Construct arguments for add_fourier_noise
idx = (1, 2)  # tuple of two ints for indexing into frequency domain

# Create a small batch of images: [B, C, H, W]
# Using distinct sizes: batch=2, channels=3, height=4, width=5
images = torch.rand(2, 3, 4, 5, dtype=torch.float32)

# Norm for the noise
norm = 0.1

# Optional size parameter - can be None or a tuple of (H, W)
size = None

ARGS = {
    "idx": idx,
    "images": images,
    "norm": norm,
    "size": size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch, tqdm — 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
cat Ran
ran, honoured the contract we drafted
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("23fe8c66119c4364")
How it ran

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

estimate_id Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("830b04cd7018c7f6")
How it ran

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

landscape1d Ran
ran; our contract draft was wrong, not the code
brando90/ultimate-anatome/anatome/landscape.py
code served (permissive licence) · get_code("5193c5e04cb818a8")
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 tuple, ndim 2, dtype family float; no output shape is stated. 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 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
from torch import nn, Tensor
from typing import Tuple, Callable

torch.manual_seed(0)

# Build a minimal model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x: Tensor) -> Tensor:
        x = self.linear1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

# Create tiny data: batch size 2, input dim 4, output dim 2
input_tensor = torch.randn(2, 4)
target_tensor = torch.randn(2, 2)
data = (input_tensor, target_tensor)

# Define a simple criterion
def criterion(output: Tensor, target: Tensor) -> Tensor:
    return torch.nn.functional.mse_loss(output, target)

# Define x_range and step_size
x_range = (-1.0, 1.0)
step_size = 0.5

ARGS = {
    "model": model,
    "data": data,
    "criterion": criterion,
    "x_range": x_range,
    "step_size": step_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, 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
landscape2d Ran
ran; our contract draft was wrong, not the code
brando90/ultimate-anatome/anatome/landscape.py
code served (permissive licence) · get_code("bc2b6fecc0fce1d0")
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 tuple, ndim 3, 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
from torch import nn
import torch.nn.functional as F

torch.manual_seed(0)

# Build a minimal model that works with the function
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
    
    def forward(self, x):
        x = F.relu(self.linear1(x))
        x = self.linear2(x)
        return x

model = SimpleModel()

# Create tiny data: batch_size=2, input_dim=4, output_dim=2
inputs = torch.randn(2, 4)
targets = torch.randint(0, 2, (2,))
data = (inputs, targets)

# Define a criterion function
def criterion(output, target):
    return F.cross_entropy(output, target)

# Define ranges and step sizes
x_range = (-1.0, 1.0)
y_range = (-1.0, 1.0)
step_size = 0.5

# We need to define _evaluate since it's called but not provided
def _evaluate(model, data, criterion):
    inputs, targets = data
    outputs = model(inputs)
    loss = criterion(outputs, targets)
    return loss

# Monkey-patch _evaluate into the module scope for the function to use
import sys
current_module = sys.modules[__name__]
setattr(current_module, '_evaluate', _evaluate)

ARGS = {
    'model': model,
    'data': data,
    'criterion': criterion,
    'x_range': x_range,
    'y_range': y_range,
    'step_size': step_size
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 3, 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
shuffle Ran
ran, honoured the contract we drafted
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("50810c2ae403377b")
How it ran

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

standardize Ran
ran, honoured the contract we drafted
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("be4cb75958ddc1e9")
How it ran

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

twoNN Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("8b3ff5a491d4198d")
How it ran

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

fourier_map Not yet run
unverified (harvested, no recorded run)
brando90/ultimate-anatome/anatome/fourier.py
code served (permissive licence) · get_code("428ae24350e49c39")
How it ran

no invocation recorded.

id_correlation Not yet run
unverified (harvested, no recorded run)
lorenzobasile/IDCorrelation/utils/metrics.py
pointer only (licence: NONE) · get_code("73a8d9c0335c1058")
How it ran

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

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, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 0 later) and 7 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
add_fourier_noise 1 other paper
1 dated earlier
2108.01661 (2021-08)
Grounding Representation Similarity with Statistical Testing
fourier_map 1 other paper
1 dated earlier
2108.01661 (2021-08)
Grounding Representation Similarity with Statistical Testing
landscape1d 1 other paper
1 dated earlier
2108.01661 (2021-08)
Grounding Representation Similarity with Statistical Testing
landscape2d 1 other paper
1 dated earlier
2108.01661 (2021-08)
Grounding Representation Similarity with Statistical Testing

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

To gain insight into the mechanisms behind machine learning methods, it is crucial to establish connections among the features describing data points. However, these correlations often exhibit a high-dimensional and strongly nonlinear nature, which makes them challenging to detect using standard methods. This paper exploits the entanglement between intrinsic dimensionality and correlation to propose a metric that quantifies the (potentially nonlinear) correlation between high-dimensional manifolds. We first validate our method on synthetic data in controlled environments, showcasing its advantages and drawbacks compared to existing techniques. Subsequently, we extend our analysis to large-scale applications in neural network representations. Specifically, we focus on latent representations of multimodal data, uncovering clear correlations between paired visual and textual embeddings, whereas existing methods struggle significantly in detecting similarity. Our results indicate the presence of highly nonlinear correlation patterns between latent manifolds.

For agents

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

get_harvested_code_for_paper("2406.15812")
get_code_for_paper("2406.15812")
have("2406.15812")

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

curl https://syntology.ai/api/ran/2406.15812.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.15812.svg)](https://syntology.ai/paper/2406.15812)

Connect an agent — have() is free.