SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2207.14812 · 2022

GLEAN: Generative Latent Bank for Image Super-Resolution and Beyond

arXiv · PDF · Open in the Atlas

Code that ran

7 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 7 have no contract verdict; 0 failures of our instrument; 2 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/2207.14812.json · badge: /api/ran/2207.14812.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
open-mmlab/mmediting canonical (classifier label)
classifier basis not recorded on the link
7 ran · 2 no recorded run
FunctionStatusWhere it lives
collate_metrics Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/utils/modelindex.py
code served (permissive licence) · get_code("54b6d9dc8ce868bb")
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

torch.manual_seed(0)

keys = [
    'Model',
    'Accuracy',
    'F1 Score',
    'Dataset',
    'Precision',
    'Training Resources',
    'Recall',
    'Download'
]

ARGS = {
    'keys': keys
}
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
convert_data_config Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/create_ceph_configs.py
code served (permissive licence) · get_code("2c460a151351c467")
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 os.path as osp
from copy import deepcopy

ARGS = {
    'data_cfg': {
        'dataset': {
            'type': 'ImageNet',
            'data_root': 'data/imagenet',
            'pipeline': [
                {
                    'type': 'mmcls.LoadImageFromFile'
                },
                {
                    'type': 'LoadMask',
                    'mask_config': {
                        'mask_list_file': 'data/masks/mask_list.txt'
                    }
                },
                {
                    'type': 'RandomLoadResizeBg',
                    'bg_dir': 'data/backgrounds'
                },
                {
                    'type': 'CompositeFg',
                    'fg_dirs': ['data/foregrounds/fg1', 'data/foregrounds/fg2'],
                    'alpha_dirs': ['data/alphas/alpha1', 'data/alphas/alpha2']
                }
            ]
        }
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
filter_metric Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/metric_mapping.py
code served (permissive licence) · get_code("0ce4b91d8aca61e4")
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

torch.manual_seed(0)

# metric_mapping: dict where keys are metric names (strings) and values are metric_info (can be any object)
metric_mapping = {
    "accuracy": {"type": "classification", "higher_is_better": True},
    "loss": {"type": "regression", "higher_is_better": False},
    "f1_score": {"type": "classification", "higher_is_better": True}
}

# summary_data: dict where values are dicts that may contain metric names as keys
summary_data = {
    "epoch_1": {
        "accuracy": 0.95,
        "loss": 0.05,
        "other_info": "some data"
    },
    "epoch_2": {
        "accuracy": 0.97,
        "loss": 0.03
    },
    "epoch_3": {
        "f1_score": 0.92,
        "other_metric": 0.88
    }
}

ARGS = {
    "metric_mapping": metric_mapping,
    "summary_data": summary_data
}
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
found_table Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/utils/modelindex.py
code served (permissive licence) · get_code("8e5814637d5291d6")
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 bool, ndim 0, dtype family float, 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.

# Analyzing the function `found_table(lines, i)`:
# - `lines` is a list of strings (table lines in markdown format)
# - `i` is an integer index
# The function checks if a markdown table is found at a given position

ARGS = {
    "lines": [
        "some header line",
        "SKIP THIS TABLE marker in line before table",
        "| column1 | column2 |",
        "| :--- | ---: |",
        "| value1 | value2 |",
        "some footer",
    ],
    "i": 2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, 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
start_from_proc Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/job_watcher.py
code served (permissive licence) · get_code("4214819cf4b6bd8d")
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.

class MockProc:
    def read(self):
        # Simulating proc output with 3 lines per task (config, blank, job_id)
        return """
/path/to/config1.yaml
some_output
Job ID: 12345
/path/to/config2.yaml
some_output
Job ID: 67890
""".strip()

root = "root_directory"
proc = MockProc()

ARGS = {
    "root": root,
    "proc": proc,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
update_intervals Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/create_ceph_configs.py
code served (permissive licence) · get_code("4bbabfed79ffbff3")
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 argparse

class Args:
    def __init__(self, iters=None):
        self.iters = iters

config = {
    'train_cfg': {
        'type': 'IterBasedTrainLoop',
        'max_iters': 100,
        'val_interval': 20
    },
    'default_hooks': {
        'logger': {
            'type': 'LoggerHook',
            'interval': 10
        },
        'checkpoint': {
            'type': 'CheckpointHook',
            'interval': 6
        }
    }
}

args = Args(iters=150)

ARGS = {
    'config': config,
    'args': args
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
update_md Ran
ran on a synthesized input
open-mmlab/mmediting/.dev_scripts/update_config_readme.py
code served (permissive licence) · get_code("185156d990381282")
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 bool, ndim 0, dtype family float, 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 os
import tempfile
import shutil

# Create a temporary directory structure with test files
temp_dir = tempfile.mkdtemp()
config_subdir = os.path.join(temp_dir, 'configs')
os.makedirs(config_subdir, exist_ok=True)

# Create some dummy config files
config_files = ['model_a.py', 'model-b.py', 'train.py']
for config_file in config_files:
    with open(os.path.join(config_subdir, config_file), 'w') as f:
        f.write('# dummy config')

# Create README.md with references to config files
readme_content = """# Test README

See [model_a](configs/model_a.py) for details.
See [model-b](configs/model-b.py) for more.
See [train](configs/train.py) for training.
"""

readme_path = os.path.join(config_subdir, 'README.md')
with open(readme_path, 'w', encoding='utf-8') as f:
    f.write(readme_content)

# Create README_zh-CN.md with references to config files
readme_zh_content = """# 测试 README

参见 [model_a](configs/model_a.py) 了解详情。
参见 [model-b](configs/model-b.py) 获取更多信息。
参见 [train](configs/train.py) 进行训练。
"""

readme_zh_path = os.path.join(config_subdir, 'README_zh-CN.md')
with open(readme_zh_path, 'w', encoding='utf-8') as f:
    f.write(readme_zh_content)

# The md_file argument should point to one of the README files
md_file_to_test = readme_path

ARGS = {
    'md_file': md_file_to_test
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, ndim 0, dtype family float, 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
check_exclude Not yet run
unverified (harvested, no recorded run)
open-mmlab/mmediting/.dev_scripts/update_ut.py
code served (permissive licence) · get_code("eda5ede48a264b22")
How it ran

no invocation recorded.

create_train_job_batch Not yet run
unverified (harvested, no recorded run)
open-mmlab/mmediting/.dev_scripts/train_benchmark.py
code served (permissive licence) · get_code("cee4c863ad3de58e")
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 9 distinct functions with an identity, 9 are also present in other papers whose code we hold (9 with at least one paper dated earlier, 0 the same month, 0 later) and 0 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
check_exclude 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
collate_metrics 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
convert_data_config 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
create_train_job_batch 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
filter_metric 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
found_table 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
start_from_proc 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
update_intervals 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks
update_md 1 other paper
1 dated earlier
1905.02716 (2019-05)
EDVR: Video Restoration with Enhanced Deformable Convolutional Networks

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 show that pre-trained Generative Adversarial Networks (GANs) such as StyleGAN and BigGAN can be used as a latent bank to improve the performance of image super-resolution. While most existing perceptual-oriented approaches attempt to generate realistic outputs through learning with adversarial loss, our method, Generative LatEnt bANk (GLEAN), goes beyond existing practices by directly leveraging rich and diverse priors encapsulated in a pre-trained GAN. But unlike prevalent GAN inversion methods that require expensive image-specific optimization at runtime, our approach only needs a single forward pass for restoration. GLEAN can be easily incorporated in a simple encoder-bank-decoder architecture with multi-resolution skip connections. Employing priors from different generative models allows GLEAN to be applied to diverse categories (\eg~human faces, cats, buildings, and cars). We further present a lightweight version of GLEAN, named LightGLEAN, which retains only the critical components in GLEAN. Notably, LightGLEAN consists of only 21% of parameters and 35% of FLOPs while achieving comparable image quality. We extend our method to different tasks including image colorization and blind image restoration, and extensive experiments show that our proposed models perform favorably in comparison to existing methods. Codes and models are available at https://github.com/open-mmlab/mmediting.

For agents

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

get_harvested_code_for_paper("2207.14812")
get_code_for_paper("2207.14812")
have("2207.14812")

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

curl https://syntology.ai/api/ran/2207.14812.json

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

[![Syntology run record](https://syntology.ai/api/ran/2207.14812.svg)](https://syntology.ai/paper/2207.14812)

Connect an agent — have() is free.