SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2411.01800 · NeurIPS · 2024

Expanding Sparse Tuning for Low Memory Usage

Xiangyang Ji, Qingming Huang, Junshu Sun, Shuhui Wang, Shufan Shen

arXiv · PDF · Open in the Atlas

Code that ran

12 of 13 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 1 violated it and 1 has no contract verdict; 9 failures 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/2411.01800.json · badge: /api/ran/2411.01800.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
ssfgunner/SNELL canonical (classifier label)
classifier basis: census
11 ran · 1 no recorded run
ssfgunner/snell canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
SNELL Ran
ran on a synthesized input · constructed an object; no result was computed
ssfgunner/snell/model/vision_transformer_timm.py
code served (permissive licence) · get_code("c7342ee375632496")
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, SNELL: 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)

ARGS = {
    "in_features": 4,
    "out_features": 3,
    "num_Hs": 2,
    "init_thres": 0.5,
    "q_dim": 16,
    "norm_p": 2,
    "SCALE_FACTOR_fc": 0.1,
    "enable_bias": True,
}
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 SNELL — 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 SNELL 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
build_optimizer Ran
ran; our contract draft was wrong, not the code
ssfgunner/SNELL/lib/utils.py
code served (permissive licence) · get_code("dae3c8f086c47ae7")
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 AdamWCustomized. 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
from torch import optim

torch.manual_seed(0)

# Create a minimal model with named parameters
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 3)
        self.linear2 = nn.Linear(3, 2)
        self.norm = nn.BatchNorm1d(3)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.norm(x)
        x = self.linear2(x)
        return x
    
    def no_weight_decay(self):
        return set()
    
    def no_weight_decay_keywords(self):
        return {}
    
    def high_lr_keywords(self):
        return {}

# Create a config object with required attributes
class Config:
    def __init__(self):
        self.opt = 'adamw'
        self.lr = 0.001
        self.weight_decay = 0.05
        self.opt_eps = 1e-8

config = Config()
model = SimpleModel()

ARGS = {
    'config': config,
    'model': model,
    'momentum': 0.9
}
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 AdamWCustomized — a AdamWCustomized output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a AdamWCustomized 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
build_transform Ran
ran; our contract draft was wrong, not the code
ssfgunner/SNELL/lib/datasets.py
code served (permissive licence) · get_code("77106d449591afd8")
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 Compose. 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 math
from argparse import Namespace

torch.manual_seed(0)

class Args:
    def __init__(self):
        self.no_aug = False
        self.input_size = 224
        self.train_interpolation = 'bilinear'
        self.direct_resize = False
        self.img_size = 224
        self.inception = False

ARGS = {
    'is_train': True,
    'args': Args()
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
timm, torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Compose — a Compose output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Compose 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
check_keywords_in_name Ran
ran, violated the contract we drafted
ssfgunner/SNELL/lib/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from leaplabthu/inline, is pointer only (licence: NONE) · get_code("b0a5beb34716d5a6")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from leaplabthu/inline, and GitHub detected no licence in that repository 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.

checkpoint_filter_fn Ran
ran; our contract draft was wrong, not the code
ssfgunner/SNELL/model/vision_transformer_timm.py
code served (permissive licence) · get_code("6d8a1f8a0697180f")
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 dict. 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
import math

torch.manual_seed(0)

# Build a minimal model that checkpoint_filter_fn expects
class PatchEmbed(nn.Module):
    def __init__(self):
        super().__init__()
        # proj weight shape: (out_channels, in_channels, kernel_h, kernel_w)
        self.proj = nn.Conv2d(3, 16, kernel_size=4, stride=4)
        self.grid_size = [8, 8]  # 64x64 input -> 16x16 patches

class TinyModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.patch_embed = PatchEmbed()
        # pos_embed shape: (1, num_patches + num_tokens, embed_dim)
        # 64 patches (8x8 grid) + 1 cls token = 65 tokens
        self.pos_embed = nn.Parameter(torch.randn(1, 65, 12))
        self.num_tokens = 1

model = TinyModel()

# Create a state_dict to pass to checkpoint_filter_fn
state_dict = {
    'model': {
        'patch_embed.proj.weight': torch.randn(16, 3, 4, 4),
        'pos_embed': torch.randn(1, 65, 12),
        'other_param': torch.randn(2, 3),
    }
}

ARGS = {
    'state_dict': state_dict,
    'model': model,
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
drop_path Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ssfgunner/SNELL/model/utils.py
code served (permissive licence) · get_code("39eace7e2822504f")
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) and returned Tensor, ndim 4, dtype family float, shape (2, 4, 3, 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 torch

torch.manual_seed(0)

# Create a small tensor for testing
# Shape: (batch_size=2, channels=4, height=3, width=3) - 4D tensor
x = torch.randn(2, 4, 3, 3, dtype=torch.float32)

ARGS = {
    "x": x,
    "drop_prob": 0.5,
    "training": True,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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, 4, 3, 3) — shape (2, 4, 3, 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
get_structured_param_num Ran
ran, honoured the contract we drafted
ssfgunner/SNELL/engine.py
code served (permissive licence) · get_code("3fb03ad9c2f6f635")
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 int, ndim 0, dtype family int, shape () 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

torch.manual_seed(0)

ARGS = {
    'structured_type': 'lora',
    'in_dim': 768,
    'out_dim': 768,
    'low_rank_dim': 8
}
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 int, ndim 0, dtype family int, shape () — shape () 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
resize_pos_embed Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ssfgunner/SNELL/model/vision_transformer_timm.py
code served (permissive licence) · get_code("191b2735955d18f4")
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 3, dtype family float, shape (1, 17, 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 math

torch.manual_seed(0)

# The function resizes positional embeddings from one shape to another
# posemb: original position embeddings tensor (batch_size, num_tokens, embedding_dim)
# posemb_new: target position embeddings tensor with desired shape
# num_tokens: number of special tokens (like cls token) to preserve
# gs_new: target grid size (height, width) for the spatial part

# Create small test tensors
embedding_dim = 4
num_special_tokens = 1
old_grid_size = 3  # 3x3 = 9 spatial tokens
old_total_tokens = num_special_tokens + (old_grid_size * old_grid_size)  # 1 + 9 = 10

# Original position embedding: (batch=1, num_tokens=10, embedding_dim=4)
posemb = torch.randn(1, old_total_tokens, embedding_dim)

# New position embedding target shape
new_grid_size = 4  # 4x4 = 16 spatial tokens
new_total_tokens = num_special_tokens + (new_grid_size * new_grid_size)  # 1 + 16 = 17
posemb_new = torch.randn(1, new_total_tokens, embedding_dim)

# Grid size for new embedding (4x4 spatial grid)
gs_new = [4, 4]

ARGS = {
    'posemb': posemb,
    'posemb_new': posemb_new,
    'num_tokens': num_special_tokens,
    'gs_new': gs_new,
}
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 3, dtype family float, shape (1, 17, 4) — shape (1, 17, 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: none by the harness; any seed is in the setup block
set_weight_decay Ran
ran; our contract draft was wrong, not the code
ssfgunner/SNELL/lib/utils.py
code served (permissive licence) · get_code("b3b2e6a0853dff09")
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 list, ndim 1, 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

torch.manual_seed(0)

# Build a small model with various parameter types
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        # Regular weight parameters (2D - will have decay)
        self.linear1 = nn.Linear(4, 3)
        # Bias parameters (1D - no decay)
        self.linear2 = nn.Linear(3, 2)
        # Parameters that match skip_keywords
        self.norm_weight = nn.Parameter(torch.randn(2, 2))
        # Parameters that match high_lr_keywords
        self.head_classifier = nn.Parameter(torch.randn(2, 2))
        # Parameter matching skip_list
        self.special_param = nn.Parameter(torch.randn(5))

model = SimpleModel()

# Construct arguments
ARGS = {
    'model': model,
    'skip_list': ('special_param',),
    'skip_keywords': ('norm',),
    'low_keywords': (),
    'high_lr_keywords': ('head',),
    'high_lr_num': 0.001
}
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 list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
trunc_normal_ Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ssfgunner/SNELL/model/utils.py
code served (permissive licence) · get_code("02566da69866c48c")
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) and returned Tensor, ndim 2, dtype family float, shape (2, 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 math

torch.manual_seed(0)

# Create a small tensor to be filled with truncated normal values
tensor = torch.empty(2, 4)

# Define arguments for trunc_normal_
ARGS = {
    'tensor': tensor,
    'mean': 0.0,
    'std': 1.0,
    'a': -2.0,
    'b': 2.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 (2, 4) — shape (2, 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: none by the harness; any seed is in the setup block
window_partition Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ssfgunner/SNELL/model/swin_timm.py
pointer only (licence: MIT); the stored copy get_code serves, from heekhero/DTL, is pointer only (licence: NONE) · get_code("993ae96666b00cb5")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from heekhero/DTL, and GitHub detected no licence in that repository 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.

window_reverse Ran
ran; our contract draft was wrong, not the code
ssfgunner/SNELL/model/swin_timm.py
pointer only (licence: MIT); the stored copy get_code serves, from heekhero/DTL, is pointer only (licence: NONE) · get_code("609922bd93c75117")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from heekhero/DTL, and GitHub detected no licence in that repository 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.

checkpoint_filter_fn Not yet run
unverified (harvested, no recorded run)
ssfgunner/SNELL/model/convnext_timm.py
code served (permissive licence) · get_code("7619bf7f57c5d2b6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 4 the same month, 6 later) and 5 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
drop_path 224 other papers
186 dated earlier, 6 the same month, 24 later, 7 same year, order not determinable, 1 undated · 38 only as an assembly brick
2003.12039 (2020-03)
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow
2007.09384 (2020-07)
Multi-Scale Positive Sample Refinement for Few-Shot Object Detection
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
and 221 more not named here
trunc_normal_ 160 other papers
127 dated earlier, 4 the same month, 27 later, 2 same year, order not determinable · 36 only as an assembly brick
1911.10194 (2019-11)
Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation
2005.10242 (2020-05)
Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere
2007.12911 (2020-07)
Tighter risk certificates for neural networks
and 157 more not named here
checkpoint_filter_fn 37 other papers
29 dated earlier, 1 the same month, 5 later, 2 same year, order not determinable · 1 only as an assembly brick
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2112.07658 (2021-12)
AdaViT: Adaptive Tokens for Efficient Vision Transformer
and 34 more not named here
check_keywords_in_name 31 other papers
25 dated earlier, 4 later, 2 same year, order not determinable
2103.14030 (2021-03)
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
2106.03650 (2021-06)
Shuffle Transformer: Rethinking Spatial Shuffle for Vision Transformer
2106.03746 (2021-06)
Efficient Training of Visual Transformers with Small Datasets
and 28 more not named here
window_partition 31 other papers
21 dated earlier, 1 the same month, 5 later, 4 same year, order not determinable
2111.11418 (2021-11)
MetaFormer Is Actually What You Need for Vision
2203.12119 (2022-03)
Visual Prompt Tuning
2204.03645 (2022-04)
DaViT: Dual Attention Vision Transformers
and 28 more not named here
resize_pos_embed 20 other papers
18 dated earlier, 2 later · 1 only as an assembly brick
2105.12723 (2021-05)
Nested Hierarchical Transformer: Towards Accurate, Data-Efficient and Interpretable Visual Understanding
2202.07800 (2022-02)
Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
and 17 more not named here
window_reverse 6 other papers
6 dated earlier
2203.12119 (2022-03)
Visual Prompt Tuning
2210.08823 (2022-10)
Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning
Guo_Improving_Robustness_of_Vision_Transformers_by_Reducing_Sensitivity_To_Patch_CVPR_2023_paper (2023 (year only))
Improving Robustness of Vision Transformers by Reducing Sensitivity to Patch Corruptions
and 3 more not named here
checkpoint_filter_fn 3 other papers
3 dated earlier
2210.08823 (2022-10)
Scaling & Shifting Your Features: A New Baseline for Efficient Model Tuning
2307.11342 (2023-07)
Tuning Pre-trained Model via Moment Probing
2312.10136 (2023-12)
Gradient-based Parameter Selection for Efficient Fine-Tuning

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

Parameter-efficient fine-tuning (PEFT) is an effective method for adapting pretrained vision models to downstream tasks by tuning a small subset of parameters. Among PEFT methods, sparse tuning achieves superior performance by only adjusting the weights most relevant to downstream tasks, rather than densely tuning the whole weight matrix. However, this performance improvement has been accompanied by increases in memory usage, which stems from two factors, i.e., the storage of the whole weight matrix as learnable parameters in the optimizer and the additional storage of tunable weight indexes. In this paper, we propose a method named SNELL (Sparse tuning with kerNELized LoRA) for sparse tuning with low memory usage. To achieve low memory usage, SNELL decomposes the tunable matrix for sparsification into two learnable low-rank matrices, saving from the costly storage of the whole original matrix. A competition-based sparsification mechanism is further proposed to avoid the storage of tunable weight indexes. To maintain the effectiveness of sparse tuning with low-rank matrices, we extend the low-rank decomposition by applying nonlinear kernel functions to the whole-matrix merging. Consequently, we gain an increase in the rank of the merged matrix, enhancing the ability of SNELL in adapting the pre-trained models to downstream tasks. Extensive experiments on multiple downstream tasks show that SNELL achieves state-of-the-art performance with low memory usage, endowing PEFT with sparse tuning to large-scale models. Codes are available at https://github.com/ssfgunner/SNELL.

For agents

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

get_harvested_code_for_paper("2411.01800")
get_code_for_paper("2411.01800")
have("2411.01800")

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

curl https://syntology.ai/api/ran/2411.01800.json

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

[![Syntology run record](https://syntology.ai/api/ran/2411.01800.svg)](https://syntology.ai/paper/2411.01800)

Connect an agent — have() is free.