SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.05052 · 2026

DeepWeightFlow: Re-Basined Flow Matching for Generating Neural Network Weights

Saumya Gupta, Zohair Shafi, Scott Biggs, Maritz Laber, Ayan Paul

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 0 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/2601.05052.json · badge: /api/ran/2601.05052.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
NNeuralDynamics/DeepWeightFlow community
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
Bunch Ran
ran on a synthesized input · constructed an object; no result was computed
NNeuralDynamics/DeepWeightFlow/flowmatching/flow_matching.py
code served (permissive licence) · get_code("0f6f0ee897d5df44")
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 an instance of its own class, Bunch: the run constructed the object and did not call it, so no result was computed. 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 Bunch class simply takes arbitrary keyword arguments and stores them
# in its __dict__. There are no parameters to construct for execution,
# but we can demonstrate it by creating a dict with some example kwargs.

ARGS = {
    'name': 'example_bunch',
    'value': 42,
    'tensor': torch.randn(2, 4),
    'list': [1, 2, 3],
    'nested': {'key': 'value'}
}
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 Bunch — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Bunch 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
FlowMatching Ran
ran on a synthesized input · constructed an object; no result was computed
NNeuralDynamics/DeepWeightFlow/flowmatching/flow_matching.py
code served (permissive licence) · get_code("04d0b153d7c9f009")
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 an instance of its own class, FlowMatching: the run constructed the object and did not call it, so no result was computed. 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
import torch.nn as nn

torch.manual_seed(0)

# Build a tiny model that takes (xt, t) and outputs a prediction
class TinyFlowModel(nn.Module):
    def __init__(self, input_dim=4, output_dim=4):
        super().__init__()
        self.input_dim = input_dim
        self.output_dim = output_dim
        self.fc1 = nn.Linear(input_dim + 1, 8)
        self.fc2 = nn.Linear(8, output_dim)
    
    def forward(self, xt, t):
        # xt: (batch, 4), t: (batch, 1)
        combined = torch.cat([xt, t], dim=1)
        h = torch.relu(self.fc1(combined))
        out = self.fc2(h)
        return out

# Build tiny dataloaders as lists of batches
# Each batch is a tuple (data, dummy_label)
def make_tiny_loader(batch_size=2, dim=4, num_batches=2):
    batches = []
    for _ in range(num_batches):
        x = torch.randn(batch_size, dim)
        y = torch.randn(batch_size, 1)
        batches.append((x, y))
    return batches

sourceloader = make_tiny_loader(batch_size=2, dim=4, num_batches=2)
targetloader = make_tiny_loader(batch_size=2, dim=4, num_batches=2)

model = TinyFlowModel(input_dim=4, output_dim=4)

ARGS = {
    'sourceloader': sourceloader,
    'targetloader': targetloader,
    'model': model,
    'mode': 'velocity',
    't_dist': 'uniform',
    'device': torch.device('cpu'),
    'normalize_pred': False,
    'geometric': False,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, tqdm — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type FlowMatching — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a FlowMatching 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

Also present in other papers' code

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

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

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

Building efficient and effective generative models for neural network weights has been a research focus of significant interest that faces challenges posed by the high-dimensional weight spaces of modern neural networks and their symmetries. Several prior generative models are limited to generating partial neural network weights, particularly for larger models, such as ResNet and ViT. Those that do generate complete weights struggle with generation speed or require finetuning of the generated models. In this work, we present Deep-WeightFlow, a Flow Matching model that operates directly in weight space to generate diverse and highaccuracy neural network weights for a variety of architectures, neural network sizes, and data modalities. The neural networks generated by DeepWeightFlow do not require fine-tuning to perform well and can scale to large networks. We apply Git Re-Basin and TransFusion for neural network canonicalization in the context of generative weight models to account for the impact of neural network permutation symmetries and to improve generation efficiency for larger model sizes. The generated networks excel at transfer learning, and ensembles of hundreds of neural networks can be generated in minutes, far exceeding the efficiency of diffusion-based methods. DeepWeightFlow models pave the way for more efficient and scalable generation of diverse sets of neural networks.

For agents

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

get_harvested_code_for_paper("2601.05052")
get_code_for_paper("2601.05052")
have("2601.05052")

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

curl https://syntology.ai/api/ran/2601.05052.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.05052.svg)](https://syntology.ai/paper/2601.05052)

Connect an agent — have() is free.