SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.11906 · CVPR · 2023

Solving Oscillation Problem in Post-Training Quantization Through a Theoretical Perspective

Rui Wang, Shilei Wen, Rongrong Ji, Fei Chao, Xiawu Zheng, Conv, Yuexiao Ma, Huixia Li, Xuefeng Xiao, Xin Pan, Qmodule 𝟐, Qmodule 𝟑, and 3 more

arXiv · PDF · Open in the Atlas

Code that ran

17 of 17 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, 3 violated it and 6 have no contract verdict; 8 failures of our instrument; 0 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2303.11906.json · badge: /api/ran/2303.11906.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
bytedance/mrecg canonical (classifier label)
classifier basis not recorded on the link
17 ran · 0 no recorded run
FunctionStatusWhere it lives
layer_has_weights Ran
ran, violated the contract we drafted
bytedance/mrecg/MRECG.py
code served (permissive licence) · get_code("734e5f593576355a")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned bool, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-05T00:24:11Z). 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)

# Build a small modules dict with some Conv2d and Linear layers
modules = {
    'layer1': nn.Conv2d(3, 16, kernel_size=3),
    'layer2': nn.Linear(10, 5),
    'layer3': nn.ReLU(),  # Not in _WEIGHTS_MODULE_TYPE
    'layer4': nn.Conv2d(16, 32, kernel_size=3),
}

# Test case 1: nodes that include weighted layers
nodes_with_weights = ['layer1', 'layer3']

# Test case 2: nodes that don't include weighted layers
nodes_without_weights = ['layer3']

# Use the first test case for ARGS
ARGS = {
    'nodes': nodes_with_weights,
    'modules': modules,
}
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 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
accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
bytedance/mrecg/utils/evaluate.py
code served (permissive licence) · get_code("b0f936d4d6ae3b8c")
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 list, ndim 2, 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

torch.manual_seed(0)

# Create output tensor: batch_size=2, num_classes=4
# This represents logits from a model
output = torch.randn(2, 4)

# Create target tensor: batch_size=2
# These are class indices (0-3 for 4 classes)
target = torch.tensor([1, 3])

# topk parameter: tuple of k values to compute accuracy for
topk = (1, 2)

ARGS = {
    "output": output,
    "target": target,
    "topk": topk,
}
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 list, ndim 2, 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
adjust_ws_gs_comp Ran
ran on a synthesized input
bytedance/mrecg/models/regnet.py
code served (permissive licence) · get_code("776c7e0f77f8b8e1")
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 tuple, ndim 2, dtype family int; 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 numpy as np

torch.manual_seed(0)

# The function adjust_ws_gs_comp takes three parameters:
# - ws: list of widths (integers)
# - bms: list of bottleneck multipliers (floats between 0 and 1)
# - gs: list of group sizes (integers)

# Looking at the function body:
# - ws is zipped with bms to compute ws_bot
# - gs is zipped with ws_bot to compute min values
# - ws_bot is zipped with g to quantize
# - ws_bot is zipped with bms to compute final ws
# All operations are list comprehensions that iterate, so these should be lists

# Create small distinct-sized lists for testing
ws = [64, 128, 256]  # widths
bms = [0.25, 0.5, 0.75]  # bottleneck multipliers (floats)
gs = [4, 8, 16]  # group sizes

ARGS = {
    "ws": ws,
    "bms": bms,
    "gs": gs
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family int — a tuple 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
bytedance/mrecg/models/resnet.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from lixiaotong97/DSU, 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.

conv_1x1_bn Ran
ran; our contract draft was wrong, not the code
bytedance/mrecg/models/mobilenet_v2.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from zhoudaquan/rethinking_bottleneck_design, is pointer only (licence: NOASSERTION) · get_code("a0131fb70c267a9e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from zhoudaquan/rethinking_bottleneck_design, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

conv_bn Ran
ran; our contract draft was wrong, not the code
bytedance/mrecg/models/mobilenet_v2.py
code served (permissive licence) · get_code("e087cbfa1e53486d")
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 Sequential. 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 = {
    "inp": 3,
    "oup": 16,
    "stride": 1
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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)
bytedance/mrecg/models/layer/drop_path.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from fudan-zvg/SeaFormer, is pointer only (licence: NONE) · get_code("3ac6b7d76e8e3584")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from fudan-zvg/SeaFormer, 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.

evaluate Ran
ran on a synthesized input
bytedance/mrecg/utils/evaluate.py
code served (permissive licence) · get_code("088c8a23c164541f")
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 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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple model
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 10)
    
    def forward(self, x):
        return self.linear(x)

# Create a simple data loader (list of batches)
val_loader = [
    (torch.randn(2, 4), torch.randint(0, 10, (2,))),
    (torch.randn(2, 4), torch.randint(0, 10, (2,))),
    (torch.randn(2, 4), torch.randint(0, 10, (2,))),
]

model = SimpleModel()

ARGS = {
    'val_loader': val_loader,
    'model': model,
    'device': torch.device('cpu'),
    'print_freq': 1
}
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 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
get_stages_from_blocks Ran
ran on a synthesized input
bytedance/mrecg/models/regnet.py
code served (permissive licence) · get_code("cf37365adf0e1e42")
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 tuple, ndim 2, dtype family int; 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 numpy as np

# Analyze the function to understand what it needs:
# - ws: appears to be a list/sequence that gets concatenated with [0]
# - rs: appears to be a list/sequence that gets concatenated with [0]
# The function compares consecutive elements to detect stage boundaries
# It returns stage widths and stage depths

# Create simple test data
ws = [64, 64, 128, 128, 256]  # widths at each block
rs = [2, 2, 4, 4, 8]           # some values at each block

ARGS = {
    "ws": ws,
    "rs": rs,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family int — a tuple 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
is_absorbing Ran
ran, violated the contract we drafted
bytedance/mrecg/utils/fold_bn.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yfguo91/ternary-spike, is pointer only (licence: NONE) · get_code("bc1bf5bb29e60206")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yfguo91/ternary-spike, 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.

is_bn Ran
ran, violated the contract we drafted
bytedance/mrecg/utils/fold_bn.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from yfguo91/ternary-spike, is pointer only (licence: NONE) · get_code("06ddc213753098ff")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from yfguo91/ternary-spike, 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.

load_calibrate_data Ran
ran on a synthesized input
bytedance/mrecg/ptq_main.py
code served (permissive licence) · get_code("ef330de8e9d4a16f")
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 3, 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)

# Create a simple data loader: list of (batch_x, batch_y) tuples
# Each batch contains input tensors and labels
train_loader = [
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
]

cali_batchsize = 2

ARGS = {
    "train_loader": train_loader,
    "cali_batchsize": cali_batchsize,
}
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 3, 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
node2modules Ran
ran; our contract draft was wrong, not the code
bytedance/mrecg/MRECG.py
code served (permissive licence) · get_code("747681743b94e243")
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
from torch.fx import GraphModule, Node

torch.manual_seed(0)

# Create a simple nn.Module for testing
class SimpleModule(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

# Create a dictionary mapping names to modules
name2modules = {
    "conv1": nn.Conv2d(1, 2, 3),
    "linear1": nn.Linear(4, 3),
    "relu": nn.ReLU(),
    "batch_norm": nn.BatchNorm1d(4),
}

# Create mock Node objects with target attributes
class MockNode:
    def __init__(self, target):
        self.target = target

# Create a list of nodes with some targets that exist in name2modules
nodes = [
    MockNode("conv1"),
    MockNode("linear1"),
    MockNode("nonexistent_module"),
    MockNode("relu"),
    MockNode("another_nonexistent"),
    MockNode("batch_norm"),
]

ARGS = {
    "name2modules": name2modules,
    "nodes": nodes,
}
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
parse_config Ran
ran; our contract draft was wrong, not the code
bytedance/mrecg/utils/utils.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from chengtao-lv/PTQ4SAM, is pointer only (licence: NONE) · get_code("afb8f93395a730c8")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from chengtao-lv/PTQ4SAM, 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.

qnode2fpnode Ran
ran on a synthesized input
bytedance/mrecg/MRECG.py
code served (permissive licence) · get_code("012346fa47909d5e")
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

class MockNode:
    def __init__(self, name):
        self.name = name

# Create quantized modules (nodes with names)
quant_modules = [
    MockNode("layer1"),
    MockNode("layer2"),
    MockNode("layer3"),
]

# Create fp32 modules (nodes with same names)
fp32_modules = [
    MockNode("layer1"),
    MockNode("layer2"),
    MockNode("layer3"),
]

ARGS = {
    "quant_modules": quant_modules,
    "fp32_modules": fp32_modules,
}
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
quantize_float Ran
ran on a synthesized input
bytedance/mrecg/models/regnet.py
code served (permissive licence) · get_code("a7f56f01db1812a0")
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

torch.manual_seed(0)

ARGS = {
    "f": 3.7,
    "q": 0.5,
}
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
search_fold_and_remove_bn Ran
ran; our contract draft was wrong, not the code
bytedance/mrecg/utils/fold_bn.py
code served (permissive licence) · get_code("ef0fe1948e3ae202")
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 Conv2d. 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)

# Create a model that contains Conv2d followed by BatchNorm2d
# This matches the structure the function expects to process
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        # Conv2d: (in_channels=3, out_channels=4, kernel_size=3)
        self.conv1 = nn.Conv2d(3, 4, kernel_size=3, padding=1)
        # BatchNorm2d: (num_features=4)
        self.bn1 = nn.BatchNorm2d(4)
        # Another Conv2d to test the prev tracking
        self.conv2 = nn.Conv2d(4, 5, kernel_size=3, padding=1)
        # Another BatchNorm2d
        self.bn2 = nn.BatchNorm2d(5)
    
    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.conv2(x)
        x = self.bn2(x)
        return x

# Instantiate the model
model = SimpleModel()

# Initialize batch norm statistics by running in eval mode
# and ensure they have proper running_mean and running_var
model.eval()

# Optionally do a forward pass to populate running statistics
dummy_input = torch.randn(2, 3, 8, 8)
with torch.no_grad():
    model(dummy_input)

ARGS = {
    "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 Conv2d — a Conv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv2d 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

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 14 are also present in other papers whose code we hold (12 with at least one paper dated earlier, 5 the same month, 9 later) and 3 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
conv3x3 686 other papers
459 dated earlier, 10 the same month, 210 later, 7 same year, order not determinable · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
drop_path 224 other papers
75 dated earlier, 10 the same month, 132 later, 6 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
accuracy 116 other papers
66 dated earlier, 1 the same month, 47 later, 2 same year, order not determinable · 8 only as an assembly brick
1502.03044 (2015-02)
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
1610.05820 (2016-10)
Membership Inference Attacks Against Machine Learning Models
1905.03696 (2019-05)
HAWQ: Hessian AWare Quantization of Neural Networks with Mixed-Precision
and 113 more not named here
conv_1x1_bn 31 other papers
22 dated earlier, 1 the same month, 6 later, 2 same year, order not determinable · 2 only as an assembly brick
1703.07402 (2017-03)
Simple Online and Realtime Tracking with a Deep Association Metric
1801.04381 (2018-01)
MobileNetV2: Inverted Residuals and Linear Bottlenecks
1811.08383 (2018-11)
TSM: Temporal Shift Module for Efficient Video Understanding
and 28 more not named here
conv_bn 18 other papers
15 dated earlier, 1 the same month, 2 later · 1 only as an assembly brick
1611.06403 (2016-11)
Deep Outdoor Illumination Estimation
1703.07402 (2017-03)
Simple Online and Realtime Tracking with a Deep Association Metric
1801.04381 (2018-01)
MobileNetV2: Inverted Residuals and Linear Bottlenecks
and 15 more not named here
is_absorbing 4 other papers
1 dated earlier, 2 later, 1 same year, order not determinable
2106.06984 (2021-06)
A Free Lunch From ANN: Towards Efficient, Accurate Spiking Neural Networks Calibration
2308.07650 (2023-08)
EQ-Net: Elastic Quantization Neural Networks
and 1 more not named here
is_bn 4 other papers
1 dated earlier, 2 later, 1 same year, order not determinable
2106.06984 (2021-06)
A Free Lunch From ANN: Towards Efficient, Accurate Spiking Neural Networks Calibration
2308.07650 (2023-08)
EQ-Net: Elastic Quantization Neural Networks
and 1 more not named here
parse_config 3 other papers
3 later
2405.03144 (2024-05)
PTQ4SAM: Post-Training Quantization for Segment Anything
Zhang_AHCPTQ_Accurate_and_Hardware-Compatible_Post-Training_Quantization_for_Segment_Anything_Model_ICCV_2025_paper (2025 (year only))
AHCPTQ: Accurate and Hardware-Compatible Post-Training Quantization for Segment Anything Model
2505.23651 (2025-05)
Merge-Friendly Post-Training Quantization for Multi-Target Domain Adaptation
search_fold_and_remove_bn 3 other papers
2 later, 1 same year, order not determinable
Liu_PD-Quant_Post-Training_Quantization_Based_on_Prediction_Difference_Metric_CVPR_2023_paper (2023 (year only))
PD-Quant: Post-Training Quantization Based on Prediction Difference Metric
2307.10554 (2023-07)
EMQ: Evolving Training-free Proxies for Automated Mixed Precision Quantization
2312.06372 (2023-12)
Ternary Spike: Learning Ternary Spikes for Spiking Neural Networks
adjust_ws_gs_comp 1 other paper
1 dated earlier
2003.13678 (2020-03)
Designing Network Design Spaces
get_stages_from_blocks 1 other paper
1 dated earlier
2003.13678 (2020-03)
Designing Network Design Spaces
layer_has_weights 1 other paper
1 dated earlier
2203.05740 (2022-03)
QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization
node2modules 1 other paper
1 dated earlier
2203.05740 (2022-03)
QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization
quantize_float 1 other paper
1 dated earlier
2003.13678 (2020-03)
Designing Network Design Spaces

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

Post-training quantization (PTQ) is widely regarded as one of the most efficient compression methods practically, benefitting from its data privacy and low computation costs. We argue that an overlooked problem of oscillation is in the PTQ methods. In this paper, we take the initiative to explore and present a theoretical proof to explain why such a problem is essential in PTQ. And then, we try to solve this problem by introducing a principled and generalized framework theoretically. In particular, we first formulate the oscillation in PTQ and prove the problem is caused by the difference in module capacity. To this end, we define the module capacity (ModCap) under data-dependent and data-free scenarios, where the differentials between adjacent modules are used to measure the degree of oscillation. The problem is then solved by selecting top-k differentials, in which the corresponding modules are jointly optimized and quantized. Extensive experiments demonstrate that our method successfully reduces the performance drop and is generalized to different neural networks and PTQ methods. For example, with 2/4 bit ResNet-50 quantization, our method surpasses the previous state-of-the-art method by 1.9%. It becomes more significant on small model quantization, e.g. surpasses BRECQ method by 6.61% on MobileNetV2 ×0.5.

For agents

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

get_harvested_code_for_paper("2303.11906")
get_code_for_paper("2303.11906")
have("2303.11906")

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

curl https://syntology.ai/api/ran/2303.11906.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.11906.svg)](https://syntology.ai/paper/2303.11906)

Connect an agent — have() is free.