SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.04489 · ACL · 2021

Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworks

Sebastian Ruder, Mostafa Dehghani, Rabeeh Karimi Mahabadi, James Henderson

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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; 0 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 1 failure of our instrument; 1 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/2106.04489.json · badge: /api/ran/2106.04489.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
rabeehk/hyperformer marked official in Papers with Code
Papers with Code archive
4 ran · 1 no recorded run
FunctionStatusWhere it lives
AdapterHyperNet Ran
ran on a synthesized input · constructed an object; no result was computed
rabeehk/hyperformer/hyperformer/adapters/adapter_controller.py
code served (permissive licence) · get_code("c4d48dab1da65576")
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, AdapterHyperNet: 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 Config:
    def __init__(self):
        self.hidden_dim = 8
        self.train_task_embeddings = True
        self.projected_task_embedding_dim = 6
        self.task_embedding_dim = 5

config = Config()
input_dim = 3
output_dim = 4

ARGS = {
    "config": config,
    "input_dim": input_dim,
    "output_dim": output_dim
}
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 AdapterHyperNet — 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 AdapterHyperNet 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
LayerNormHyperNet Ran
ran on a synthesized input · constructed an object; no result was computed
rabeehk/hyperformer/hyperformer/adapters/adapter_controller.py
code served (permissive licence) · get_code("7929f8a733e5bfbc")
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, LayerNormHyperNet: 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 Config:
    def __init__(self):
        self.projected_task_embedding_dim = 5
        self.train_task_embeddings = True
        self.task_embedding_dim = 3
        self.input_dim = 4

config = Config()

ARGS = {
    "config": config
}
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 LayerNormHyperNet — 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 LayerNormHyperNet 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
MetaAdapterController Ran
ran on a synthesized input · constructed an object; no result was computed
rabeehk/hyperformer/hyperformer/adapters/adapter_controller.py
code served (permissive licence) · get_code("7e8ce70c56f6a890")
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, MetaAdapterController: 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 Config:
    def __init__(self):
        self.hidden_dim = 8
        self.input_dim = 16
        self.projected_task_embedding_dim = 12
        self.task_embedding_dim = 10
        self.train_task_embeddings = True
        self.reduction_factor = 2
        self.non_linearity = "relu"
        self.add_layer_norm_before_adapter = True
        self.add_layer_norm_after_adapter = True
        self.conditional_layer_norm = True
        self.device = "cpu"

config = Config()

task_embedding = torch.randn(1, config.projected_task_embedding_dim)

inputs = torch.randn(2, 16)

ARGS = {
    "config": config,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MetaAdapterController — 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 MetaAdapterController 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
linear_layer Ran
ran; our contract draft was wrong, not the code
rabeehk/hyperformer/hyperformer/adapters/adapter_controller.py
code served (permissive licence) · get_code("2abb955708c84118")
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 Linear. 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)

ARGS = {
    "input_dim": 4,
    "output_dim": 3,
    "std": 0.01,
}
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 Linear — a Linear output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Linear 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
init_linear_layer Not yet run
unverified (harvested, no recorded run)
rabeehk/hyperformer/hyperformer/adapters/adapter_controller.py
code served (permissive licence) · get_code("d4d0162fd3f6583b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 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 4 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
linear_layer
brick of this paper's assembly: helper
3 other papers
3 later
2112.06825 (2021-12)
VL-Adapter: Parameter-Efficient Transfer Learning for Vision-and-Language Tasks
2203.01104 (2022-03)
Parameter-Efficient Mixture-of-Experts Architecture for Pre-trained Language Models
2024.findings-acl.699 (2024 (year only))
SCALEARN: Simple and Highly Parameter-Efficient Task Transfer by Learning to Scale

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

State-of-the-art parameter-efficient fine-tuning methods rely on introducing adapter modules between the layers of a pretrained language model. However, such modules are trained separately for each task and thus do not enable sharing information across tasks. In this paper, we show that we can learn adapter parameters for all layers and tasks by generating them using shared hypernetworks, which condition on task, adapter position, and layer id in a transformer model. This parameter-efficient multi-task learning framework allows us to achieve the best of both worlds by sharing knowledge across tasks via hypernetworks while enabling the model to adapt to each individual task through task-specific adapters. Experiments on the well-known GLUE benchmark show improved performance in multi-task learning while adding only 0.29% parameters per task. We additionally demonstrate substantial performance improvements in few-shot domain generalization across a variety of tasks. Our code is publicly available in https://github.com/ rabeehk/hyperformer.

For agents

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

get_harvested_code_for_paper("2106.04489")
get_code_for_paper("2106.04489")
have("2106.04489")

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

curl https://syntology.ai/api/ran/2106.04489.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.04489.svg)](https://syntology.ai/paper/2106.04489)

Connect an agent — have() is free.