SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2402.03412 · ICML · 2024

See More Details: Efficient Image Super-Resolution by Experts Mining

Radu Timofte, Yulun Zhang, Eduard Zamfir, Zongwei Wu, Nancy Mehta

arXiv · PDF · Open in the Atlas

Code that ran

15 of 19 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (13 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 14 have no contract verdict; 1 failure of our instrument; 4 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/2402.03412.json · badge: /api/ran/2402.03412.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
eduardzamfir/seemoredetails canonical (classifier label)
classifier basis: classifier_2025
7 ran · 3 no recorded run
zhoushen1/MEASNet community
Papers with Code archive
8 ran · 1 no recorded run
FunctionStatusWhere it lives
EL Ran
ran on a synthesized input · constructed an object; no result was computed
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("1d1bc8a1a6d7eecd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

Expert Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("8e705ad33db3ece0")
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, Expert: 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_ch": 4,
    "low_dim": 3,
}
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 Expert — 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 Expert 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
FD Ran
ran on a synthesized input
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("b890f94263e3455b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

GatedFFN Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("66c26e9c9b1d539f")
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, GatedFFN: 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)

# Parameters for GatedFFN
in_ch = 8
mlp_ratio = 2
kernel_size = 3
act_layer = nn.GELU()

ARGS = {
    "in_ch": in_ch,
    "mlp_ratio": mlp_ratio,
    "kernel_size": kernel_size,
    "act_layer": act_layer,
}
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 GatedFFN — 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 GatedFFN 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
Layer Ran
ran on a synthesized input · constructed an object; no result was computed
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("9146fec2223e3a45")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

MEE Ran
ran on a synthesized input · constructed an object; no result was computed
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("26e700a2cf13763b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

MESE Ran
ran on a synthesized input · constructed an object; no result was computed
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("17e909eea51c6bed")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

MoEBlock Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("4f3e4c93dfd5cb1a")
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, MoEBlock: 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_ch': 4,
    'num_experts': 3,
    'topk': 2,
    'use_shuffle': False,
    'lr_space': 'linear',
    'recursive': 2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type MoEBlock — 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 MoEBlock 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
MoELayer Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("d161ce9a080bff74")
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, MoELayer: 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a simple expert module that takes inputs and k
class SimpleExpert(nn.Module):
    def __init__(self, input_dim, output_dim):
        super().__init__()
        self.linear = nn.Linear(input_dim, output_dim)
    
    def forward(self, inputs, k):
        # inputs shape: (batch, channels, height, width) or similar
        # Return same shape as inputs
        return self.linear(inputs)

# Create a simple gate module
class SimpleGate(nn.Module):
    def __init__(self, input_dim, num_experts):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_experts)
    
    def forward(self, inputs):
        # Flatten inputs to (batch, -1) for linear layer
        batch_size = inputs.shape[0]
        flat_inputs = inputs.view(batch_size, -1)
        return self.linear(flat_inputs)

# Set up dimensions
batch_size = 2
num_channels = 4
height = 3
width = 5
input_dim = num_channels * height * width
num_experts = 3

# Create expert modules
experts = [
    SimpleExpert(input_dim, input_dim)
    for _ in range(num_experts)
]

# Create gate module
gate = SimpleGate(input_dim, num_experts)

# Create test input tensor with shape (batch, channels, height, width)
inputs = torch.randn(batch_size, num_channels, height, width)

# Create k tensor (used in expert forward pass)
k = torch.tensor([1, 2])

# Build the arguments dict
ARGS = {
    'experts': experts,
    'gate': gate,
    'num_expert': 2
}
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 MoELayer — 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 MoELayer 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
Router Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("f825feff5cc9f504")
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, Router: 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
from einops.layers.torch import Rearrange

torch.manual_seed(0)

ARGS = {
    "in_ch": 4,
    "num_experts": 3,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Router — 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 Router 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
SME Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("37b23826145deabe")
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, SME: 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_ch': 4,
    'kernel_size': 11,
}
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 SME — 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 SME 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
StripedConv2d Ran
ran on a synthesized input · constructed an object; no result was computed
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("1b9acd1b10cbbbff")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

StripedConvFormer Ran
ran on a synthesized input · constructed an object; no result was computed
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("513ee53b4c559a5a")
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, StripedConvFormer: 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_ch": 4,
    "kernel_size": 3,
}
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 StripedConvFormer — 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 StripedConvFormer 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
WET Ran
ran on a synthesized input · constructed an object; no result was computed
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("4d9c76578a8d3647")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

channel_shuffle Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("80663608dbd5eff8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

FD_MEE Not yet run
unverified (harvested, no recorded run)
zhoushen1/MEASNet/MEASNet/net/FD_MEE.py
pointer only (licence: NONE) · get_code("3cb1c1cb77fa48b8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zhoushen1/MEASNet 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.

RME Not yet run
unverified (harvested, no recorded run)
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("fb65e199251d159e")
How it ran

no invocation recorded.

ResGroup Not yet run
unverified (harvested, no recorded run)
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("69415565c8c15832")
How it ran

no invocation recorded.

SeemoRe Not yet run
unverified (harvested, no recorded run)
eduardzamfir/seemoredetails/basicsr/archs/seemore_arch.py
code served (permissive licence) · get_code("ddb8a8108be5a5c7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 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 19 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

Reconstructing high-resolution (HR) images from low-resolution (LR) inputs poses a significant challenge in image super-resolution (SR). While recent approaches have demonstrated the efficacy of intricate operations customized for various objectives, the straightforward stacking of these disparate operations can result in a substantial computational burden, hampering their practical utility. In response, we introduce SeemoRe, an efficient SR model employing expert mining. Our approach strategically incorporates experts at different levels, adopting a collaborative methodology. At the macro scale, our experts address rank-wise and spatial-wise informative features, providing a holistic understanding. Subsequently, the model delves into the subtleties of rank choice by leveraging a mixture of low-rank experts. By tapping into experts specialized in distinct key factors crucial for accurate SR, our model excels in uncovering intricate intra-feature details. This collaborative approach is reminiscent of the concept of "see more", allowing our model to achieve an optimal performance with minimal computational costs in efficient settings. The source codes will be publicly made available at https://github. com/eduardzamfir/seemoredetails

For agents

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

get_harvested_code_for_paper("2402.03412")
get_code_for_paper("2402.03412")
have("2402.03412")

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

curl https://syntology.ai/api/ran/2402.03412.json

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

[![Syntology run record](https://syntology.ai/api/ran/2402.03412.svg)](https://syntology.ai/paper/2402.03412)

Connect an agent — have() is free.