SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2006.14769 · NeurIPS · 2020

Supermasks in Superposition

Vivek Ramanujan, Mitchell Wortsman, Ali Farhadi, Rosanne Liu, Jason Yosinski, Aniruddha Kembhavi, M Collective, Mohammad Rastegari, Supermask, Task

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 1 failure of our instrument; 3 no recorded run), measured 2026-11-21; 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/2006.14769.json · badge: /api/ran/2006.14769.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
RAIVNLab/supsup canonical (classifier label)
classifier basis: classifier_2025
2 ran · 2 no recorded run
dmiu-shell/deeprl-shell community
Papers with Code archive
4 ran · 1 no recorded run
FunctionStatusWhere it lives
GetSubnetContinuous Ran
ran on a synthesized input · constructed an object; no result was computed
dmiu-shell/deeprl-shell/deep_rl/shell_modules/mmn/ssmask_utils.py
code served (permissive licence) · get_code("322b181b10018053")
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, GetSubnetContinuous: 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.autograd as autograd

torch.manual_seed(0)

# Create a small scores tensor (batch_size=2, features=4)
scores = torch.randn(2, 4, requires_grad=True)

# Define threshold parameter
a = 0.0

# Construct the ARGS dictionary
ARGS = {
    'scores': scores,
    'a': a
}
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 GetSubnetContinuous — 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 GetSubnetContinuous 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
GetSubnetDiscrete Ran
ran on a synthesized input · constructed an object; no result was computed
dmiu-shell/deeprl-shell/deep_rl/shell_modules/mmn/ssmask_utils.py
code served (permissive licence) · get_code("4de2d75cb07fd186")
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, GetSubnetDiscrete: 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.autograd as autograd

torch.manual_seed(0)

# Create a small scores tensor
scores = torch.randn(2, 4, requires_grad=True)

# Define the threshold parameter
a = 0.5

ARGS = {
    'scores': scores,
    'a': a,
}
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 GetSubnetDiscrete — 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 GetSubnetDiscrete 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
MultitaskMaskLinear Ran
ran on a synthesized input · constructed an object; no result was computed
dmiu-shell/deeprl-shell/deep_rl/shell_modules/mmn/ssmask_utils.py
code served (permissive licence) · get_code("31da33c140f3a9c8")
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, MultitaskMaskLinear: 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 math
import torch
import torch.nn as nn

torch.manual_seed(0)

ARGS = {
    'in_features': 4,
    'out_features': 3,
    'discrete': True,
    'num_tasks': 2,
    'new_mask_type': 'random',
    'bias': 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 MultitaskMaskLinear — 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 MultitaskMaskLinear 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
get_subnet Ran
ran; our contract draft was wrong, not the code
RAIVNLab/supsup/models/modules.py
pointer only (licence: NONE) · get_code("a407b1a72b42c61f")
How it ran

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

mask_init Ran
ran, honoured the contract we drafted
RAIVNLab/supsup/models/modules.py
pointer only (licence: NONE) · get_code("ffe66fb2efd3fe30")
How it ran

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

mask_init Ran
ran, honoured the contract we drafted
dmiu-shell/deeprl-shell/deep_rl/shell_modules/mmn/ssmask_utils.py
code served (permissive licence) · get_code("4041eef325f5a3b2")
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 (4, 3) 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 math
import torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.weight = nn.Parameter(torch.randn(4, 3))

module = SimpleModule()

ARGS = {
    "module": module
}
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 (4, 3) — shape (4, 3) 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
MultitaskMaskConv Not yet run
unverified (harvested, no recorded run)
RAIVNLab/supsup/models/modules.py
pointer only (licence: NONE) · get_code("1d23e73da3824929")
How it ran

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

kwargs_to_cmd Not yet run
unverified (harvested, no recorded run)
RAIVNLab/supsup/experiments/GG/splitcifar100/rn18-supsup.py
pointer only (licence: NONE) · get_code("f2849754778de9a4")
How it ran

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

signed_constant Not yet run
unverified (harvested, no recorded run)
dmiu-shell/deeprl-shell/deep_rl/shell_modules/mmn/ssmask_utils.py
code served (permissive licence) · get_code("3220897509d99823")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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
mask_init
brick of this paper's assembly: helper
4 other papers
4 later
2210.10209 (2022-10)
Exclusive Supermask Subnetwork Training for Continual Learning
2401.13200 (2024-01)
Topology-aware Embedding Memory for Continual Learning on Expanding Networks
2406.07413 (2024-06)
Towards Continuous Reuse of Graph Models via Holistic Memory Diversification
and 1 more not named here

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 2026-11-21.

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 present the Supermasks in Superposition (SupSup) model, capable of sequentially learning thousands of tasks without catastrophic forgetting. Our approach uses a randomly initialized, fixed base network and for each task finds a subnetwork (supermask) that achieves good performance. If task identity is given at test time, the correct subnetwork can be retrieved with minimal memory usage. If not provided, SupSup can infer the task using gradient-based optimization to find a linear superposition of learned supermasks which minimizes the output entropy. In practice we find that a single gradient step is often sufficient to identify the correct mask, even among 2500 tasks. We also showcase two promising extensions. First, SupSup models can be trained entirely without task identity information, as they may detect when they are uncertain about new data and allocate an additional supermask for the new training distribution. Finally the entire, growing set of supermasks can be stored in a constant-sized reservoir by implicitly storing them as attractors in a fixed-sized Hopfield network.

For agents

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

get_harvested_code_for_paper("2006.14769")
get_code_for_paper("2006.14769")
have("2006.14769")

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

curl https://syntology.ai/api/ran/2006.14769.json

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

[![Syntology run record](https://syntology.ai/api/ran/2006.14769.svg)](https://syntology.ai/paper/2006.14769)

Connect an agent — have() is free.