SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1709.07871 · 2017

FiLM: Visual Reasoning with a General Conditioning Layer

Ethan Perez, Aaron Courville, Florian Strub, Vincent Dumoulin, Harm De Vries

arXiv · PDF · Open in the Atlas

Code that ran

9 of 10 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 1 failure of our instrument; 1 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/1709.07871.json · badge: /api/ran/1709.07871.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
caffeinism/film-pytorch community
Papers with Code archive
5 ran · 1 no recorded run
keonlee9420/Daft-Exprt community
Papers with Code archive
1 ran · 0 no recorded run
CPJKU/audio_conditioned_unet community
Papers with Code archive
1 ran · 0 no recorded run
ethanjperez/film marked official in Papers with Code
Papers with Code archive
1 ran · 0 no recorded run
kdaip/stabletts community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
Classifier Ran
ran on a synthesized input · constructed an object; no result was computed
caffeinism/film-pytorch/networks.py
pointer only (licence: NONE) · get_code("37dda49b6caaba23")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, caffeinism/film-pytorch 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.

FeatureExtractor Ran
ran on a synthesized input · constructed an object; no result was computed
caffeinism/film-pytorch/networks.py
pointer only (licence: NONE) · get_code("edbfcccbabd461c6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, caffeinism/film-pytorch 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.

FiLM Ran
ran on a synthesized input
keonlee9420/Daft-Exprt/model/blocks.py
code served (permissive licence) · get_code("bc5a2b1350c91ae2")
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 3, dtype family float, shape (2, 3, 4) 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 torch.nn as nn

class FiLM(nn.Module):
    """
    A Feature-wise Linear Modulation Layer from
    'FiLM: Visual Reasoning with a General Conditioning Layer'
    , extended to 'TADAM: Task dependent adaptive metric for improved few-shot learning'
    """

    def __init__(self):
        super(FiLM, self).__init__()
        self.s_gamma = nn.Parameter(torch.ones(1), requires_grad=True)
        self.s_beta = nn.Parameter(torch.ones(1), requires_grad=True)

    def forward(self, x, gammas, betas):
        """
        x -- [B, T, H]
        gammas -- [B, 1, H]
        betas -- [B, 1, H]
        """
        gammas = self.s_gamma * gammas.expand_as(x)
        betas = self.s_beta * betas.expand_as(x)
        return (gammas + 1.0) * x + betas


INIT_ARGS = {}

ARGS = {
    'x': torch.randn(2, 3, 4),
    'gammas': torch.randn(2, 1, 4),
    'betas': torch.randn(2, 1, 4),
}

if __name__ == '__main__':
    film = FiLM(**INIT_ARGS)
    output = film(**ARGS)
    print("Output shape:", output.shape)
    print("Output:", output)
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 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 4) 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)
FiLM Ran
ran on a synthesized input
CPJKU/audio_conditioned_unet/audio_conditioned_unet/network.py
code served (permissive licence) · get_code("bd2e6b10e2f47993")
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; 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 torch.nn as nn

class FiLM(nn.Module):

    def __init__(self, zdim, maskdim):
        super(FiLM, self).__init__()
        self.gamma = nn.Linear(zdim, maskdim)
        self.beta = nn.Linear(zdim, maskdim)

    def forward(self, x, z):
        gamma = self.gamma(z).unsqueeze(-1).unsqueeze(-1)
        beta = self.beta(z).unsqueeze(-1).unsqueeze(-1)
        x = gamma * x + beta
        return x


INIT_ARGS = {
    'zdim': 4,
    'maskdim': 3,
}

ARGS = {
    'x': torch.randn(2, 3, 5, 6),
    'z': torch.randn(2, 4),
}

if __name__ == '__main__':
    model = FiLM(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: {ARGS['x'].shape}")
    assert output.shape == ARGS['x'].shape, "Shape mismatch!"
    print("Success!")
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 — 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)
FiLM Ran
ran on a synthesized input
ethanjperez/film/vr/models/filmed_net.py
pointer only (licence: NOASSERTION) · get_code("b8723266db5e5556")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ethanjperez/film (NOASSERTION). 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.

FiLMBlock Ran
ran on a synthesized input
caffeinism/film-pytorch/networks.py
pointer only (licence: NONE) · get_code("32874c063a45b37a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, caffeinism/film-pytorch 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.

FiLMLayer Ran
ran on a synthesized input · constructed an object; no result was computed
kdaip/stabletts/models/estimator.py
code served (permissive licence) · get_code("04d3db3a0eb24a12")
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, FiLMLayer: 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)

class FiLMLayer(nn.Module):
    """
    Feature-wise Linear Modulation (FiLM) layer
    Reference: https://arxiv.org/abs/1709.07871
    """

    def __init__(self, in_channels, cond_channels):
        super(FiLMLayer, self).__init__()
        self.in_channels = in_channels
        self.film = nn.Conv1d(cond_channels, in_channels * 2, 1)

    def forward(self, x, c):
        gamma, beta = torch.chunk(self.film(c.unsqueeze(2)), chunks=2, dim=1)
        return gamma * x + beta


# Create arguments for FiLMLayer
in_channels = 4
cond_channels = 3

ARGS = {
    "in_channels": in_channels,
    "cond_channels": cond_channels,
}
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 FiLMLayer — 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 FiLMLayer 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
ResBlock Ran
ran on a synthesized input · constructed an object; no result was computed
caffeinism/film-pytorch/networks.py
pointer only (licence: NONE) · get_code("779ef9590306ac85")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, caffeinism/film-pytorch 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.

conv Ran
ran; our contract draft was wrong, not the code
caffeinism/film-pytorch/networks.py
pointer only (licence: NONE) · get_code("b4a8caf47616fa06")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, caffeinism/film-pytorch 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.

FiLM Not yet run
unverified (harvested, no recorded run)
caffeinism/film-pytorch/networks.py
pointer only (licence: NONE) · get_code("6d21918b730bc046")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, caffeinism/film-pytorch 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 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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

We introduce a general-purpose conditioning method for neural networks called FiLM: Feature-wise Linear Modulation. FiLM layers influence neural network computation via a simple, feature-wise affine transformation based on conditioning information. We show that FiLM layers are highly effective for visual reasoning -answering image-related questions which require a multi-step, high-level process -a task which has proven difficult for standard deep learning methods that do not explicitly model reasoning. Specifically, we show on visual reasoning tasks that FiLM layers 1) halve state-of-theart error for the CLEVR benchmark, 2) modulate features in a coherent manner, 3) are robust to ablations and architectural modifications, and 4) generalize well to challenging, new data from few examples or even zero-shot.

For agents

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

get_harvested_code_for_paper("1709.07871")
get_code_for_paper("1709.07871")
have("1709.07871")

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

curl https://syntology.ai/api/ran/1709.07871.json

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

[![Syntology run record](https://syntology.ai/api/ran/1709.07871.svg)](https://syntology.ai/paper/1709.07871)

Connect an agent — have() is free.