SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.00797 · ICML · 2026

Zero-Flow Encoders

Taiji Suzuki, Song Liu, Leyang Wang, Yakun Wang

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 (0 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/2602.00797.json · badge: /api/ran/2602.00797.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
probabilityFLOW/zfe community
Papers with Code archive
2 ran · 0 no recorded run
FunctionStatusWhere it lives
Decoder Ran
ran on a synthesized input
probabilityFLOW/zfe/models/nn.py
code served (permissive licence) · get_code("41846dfc9e040d54")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 4, dtype family float, shape (2, 1, 32, 32) 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 nn

class Decoder(nn.Module):

    def __init__(self, encx_dim=64):
        super(Decoder, self).__init__()
        self.fc_map = nn.Linear(encx_dim, 64 * 4 * 4)
        self.net = nn.Sequential(nn.ConvTranspose2d(64, 32, kernel_size=4, stride=2, padding=1), nn.BatchNorm2d(32), nn.ReLU(True), nn.ConvTranspose2d(32, 16, kernel_size=4, stride=2, padding=1), nn.BatchNorm2d(16), nn.ReLU(True), nn.ConvTranspose2d(16, 1, kernel_size=4, stride=2, padding=1), nn.Sigmoid())

    def forward(self, x):
        x = self.fc_map(x)
        x = x.view(-1, 64, 4, 4)
        img = self.net(x)
        return img


INIT_ARGS = {
    'encx_dim': 32
}

ARGS = {
    'x': torch.randn(2, 32)
}

if __name__ == '__main__':
    decoder = Decoder(**INIT_ARGS)
    output = decoder(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output min: {output.min()}, max: {output.max()}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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, 1, 32, 32) — shape (2, 1, 32, 32) 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
VectorField Ran
ran on a synthesized input
probabilityFLOW/zfe/models/nn.py
code served (permissive licence) · get_code("439020c1ecd54c21")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 2, 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
import numpy as np

INIT_ARGS = {
    "encx_dim": 4
}

# Create input tensors matching the expected shapes
batch_size = 2
y = torch.randn(batch_size, 32 * 32)  # Flattened 32x32 image
encx = torch.randn(batch_size, INIT_ARGS["encx_dim"])  # Encoded x with dim=4
x = torch.randn(batch_size, 32 * 32)  # Flattened 32x32 image
t = torch.rand(batch_size)  # Time values

ARGS = {
    "y": y,
    "encx": encx,
    "x": x,
    "t": t
}
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)

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

Flow-based methods have achieved significant success in various generative modeling tasks, capturing nuanced details within complex data distributions. However, few existing works have exploited this unique capability to resolve finegrained structural details beyond generation tasks. This paper presents a flow-inspired framework for representation learning. First, we demonstrate that a rectified flow trained using independent coupling is zero everywhere at t = 0.5 if and only if the source and target distributions are identical. We term this property the zeroflow criterion. Second, we show that this criterion can certify conditional independence, thereby extracting sufficient information from the data. Third, we translate this criterion into a tractable, simulation-free loss function that enables learning amortized Markov blankets in graphical models and latent representations in self-supervised learning tasks. Experiments on both simulated and real-world datasets demonstrate the effectiveness of our approach. The code reproducing our experiments can be found at: https: //github.com/probabilityFLOW/zfe.

For agents

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

get_harvested_code_for_paper("2602.00797")
get_code_for_paper("2602.00797")
have("2602.00797")

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

curl https://syntology.ai/api/ran/2602.00797.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.00797.svg)](https://syntology.ai/paper/2602.00797)

Connect an agent — have() is free.