SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2204.08491 · 2022

Active Learning Helps Pretrained Models Learn the Intended Task

arXiv · PDF · Open in the Atlas

Code that ran

14 of 16 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, 1 violated it and 10 have no contract verdict; 3 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/2204.08491.json · badge: /api/ran/2204.08491.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
alextamkin/active-learning-pretrained-models canonical (classifier label)
classifier basis not recorded on the link
14 ran · 2 no recorded run
FunctionStatusWhere it lives
conv1x1 Ran
ran; our contract draft was wrong, not the code
alextamkin/active-learning-pretrained-models/models/architecture.py
code served (permissive licence) · get_code("ee80cf467065076b")
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 StdConv2d. 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 = {
    'cin': 3,
    'cout': 16,
    'stride': 1,
    'bias': False,
}
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 StdConv2d — a StdConv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StdConv2d 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
conv3x3 Ran
ran; our contract draft was wrong, not the code
alextamkin/active-learning-pretrained-models/models/architecture.py
code served (permissive licence) · get_code("f01b8d3901f0b289")
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 StdConv2d. 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 = {
    'cin': 3,
    'cout': 16,
    'stride': 1,
    'groups': 1,
    'bias': False,
}
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 StdConv2d — a StdConv2d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StdConv2d 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
entropy Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/models/acquisition.py
code served (permissive licence) · get_code("0123d0a848b9b0b8")
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 1, dtype family float, shape (2,) 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)

class SimpleClassifier(nn.Module):
    def __init__(self, input_dim=4, num_classes=3):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleClassifier(input_dim=4, num_classes=3)
x = torch.randn(2, 4)

ARGS = {
    "model": model,
    "x": x
}
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 1, dtype family float, shape (2,) — shape (2,) 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
find_few_shot_subset Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/utils/multi_attr_dataset.py
code served (permissive licence) · get_code("781e1dfa44bed028")
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 Subset. 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 numpy as np

torch.manual_seed(0)

class MockMultiAttrVisionDataset(torch.utils.data.Dataset):
    """Mock dataset that implements the interface expected by find_few_shot_subset"""
    
    def __init__(self):
        # Create mock data: 6 samples with 2 attributes
        self.num_samples = 6
        self.num_attrs = 2
        
        # Mock targets: shape (6, 2) where each row is [attr0_class, attr1_class]
        self._targets = [
            torch.tensor([0, 0]),  # sample 0
            torch.tensor([0, 1]),  # sample 1
            torch.tensor([1, 0]),  # sample 2
            torch.tensor([1, 1]),  # sample 3
            torch.tensor([0, 0]),  # sample 4
            torch.tensor([1, 1]),  # sample 5
        ]
        
        # Map attribute names to indices
        self.attr_to_idx = {
            'color': 0,
            'shape': 1
        }
        
        # Map attribute names to class indices
        self.class_to_idx = {
            'color': {0: 0, 1: 1},
            'shape': {0: 0, 1: 1}
        }
        
        # List of classes per attribute
        self.attrs = ['color', 'shape']
        self.classes = {
            'color': ['red', 'blue'],
            'shape': ['circle', 'square']
        }
    
    def __getitem__(self, index):
        return {
            'image': torch.randn(3, 4),
            'target': self._targets[index]
        }
    
    def __len__(self):
        return self.num_samples
    
    def get_targets(self):
        return self._targets


# Create mock dataset instance
dataset = MockMultiAttrVisionDataset()

# Define arguments for find_few_shot_subset
ARGS = {
    'dataset': dataset,
    'attr': 'color',
    'examples_per_class': 2,
    'seed': 42
}
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 Subset — a Subset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Subset 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
get_losses_and_topks_results Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/utils/train_test_common.py
code served (permissive licence) · get_code("9e433b12009a8d58")
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. 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 simple model that takes input of shape (batch_size, num_classes) and outputs logits
class SimpleClassifier(nn.Module):
    def __init__(self, input_dim, num_classes):
        super().__init__()
        self.fc = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.fc(x)

# Set up parameters
batch_size = 2
num_classes = 4
input_dim = 3
num_attrs = 5

# Create model
model = SimpleClassifier(input_dim, num_classes)

# Create input tensor x: shape (batch_size, input_dim)
x = torch.randn(batch_size, input_dim)

# Create target tensor y: shape (batch_size, num_attrs)
# This will be indexed with attr_idx to get classes
y = torch.randint(0, num_classes, (batch_size, num_attrs))

# Attribute index to extract from y
attr_idx = 2

# Values of k for top-k evaluation
ks = (1, 2)

ARGS = {
    'model': model,
    'x': x,
    'y': y,
    'attr_idx': attr_idx,
    'ks': ks
}
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 tuple — a tuple output has no array shape of its own; none is 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
get_mixup Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/models/hyper_params.py
code served (permissive licence) · get_code("7d63034d0bfadd78")
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 float, 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 numpy as np

torch.manual_seed(0)

# The function get_mixup takes dataset_size as a parameter
# It returns 0.0 if dataset_size < 20000, else 0.1
# We need to construct a valid dataset_size argument

# Test with a small dataset size (less than 20000)
dataset_size_small = 15000

# Test with a large dataset size (>= 20000)
dataset_size_large = 25000

# We can use either one; let's use the large one to test the else branch
ARGS = {
    "dataset_size": 25000
}
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 float, 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_scaled_resolution Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/utils/datasets/metadata.py
code served (permissive licence) · get_code("f7f802d7a86c68c5")
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 1, 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

torch.manual_seed(0)

ARGS = {
    "original_resolution": (96, 96)
}
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 tuple, ndim 1, 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
get_schedule Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/models/hyper_params.py
code served (permissive licence) · get_code("2c51b4719ea4c6b9")
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 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

torch.manual_seed(0)

ARGS = {
    "dataset_size": 15000
}
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 int — 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
minimum_margin Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/models/acquisition.py
code served (permissive licence) · get_code("2c8a0fc57eee370a")
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 1, dtype family float, shape (2,) 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)

class SimpleClassifier(nn.Module):
    def __init__(self, input_dim=4, num_classes=3):
        super().__init__()
        self.linear = nn.Linear(input_dim, num_classes)
    
    def forward(self, x):
        return self.linear(x)

model = SimpleClassifier(input_dim=4, num_classes=3)
x = torch.randn(2, 4)

ARGS = {
    "model": model,
    "x": x
}
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 1, dtype family float, shape (2,) — shape (2,) 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
random Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/models/acquisition.py
code served (permissive licence) · get_code("2152d1eded7cbdc7")
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 1, dtype family float, shape (2,) 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)

# Build a minimal model - the function doesn't use it, but we need to provide it
model = nn.Linear(4, 2)

# Build a small input tensor with batch size 2 and feature dimension 4
x = torch.randn(2, 4)

ARGS = {
    "model": model,
    "x": x,
}
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 1, dtype family float, shape (2,) — shape (2,) 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
seed_dataloader_worker Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/utils/train_test_common.py
code served (permissive licence) · get_code("1b8302f01bd10a39")
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 function. 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 random
import numpy as np

# The function seed_dataloader_worker takes worker_seed as a parameter
# and returns a worker_init_fn function that can be used with PyTorch DataLoader

# Looking at the function body:
# - worker_seed is used in the returned function
# - The returned function takes worker_id as parameter
# - It sets np.random.seed(worker_seed) and random.seed(worker_seed)
# So worker_seed should be an integer

ARGS = {
    "worker_seed": 42
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
setup_logger Ran
ran on a synthesized input
alextamkin/active-learning-pretrained-models/utils/logger.py
code served (permissive licence) · get_code("12323f0640436acc")
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 Logger. 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 logging.config
import os
import tempfile

# Create a temporary directory for logging
temp_logdir = tempfile.mkdtemp()

# Create a simple args class with the required attributes
class Args:
    def __init__(self, logdir, name):
        self.logdir = logdir
        self.name = name

ARGS = {
    "args": Args(logdir=temp_logdir, name="test_run")
}
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 Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger 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
tf2th Ran
ran, violated the contract we drafted
alextamkin/active-learning-pretrained-models/models/architecture.py
code served (permissive licence) · get_code("7023eb86801bb3a3")
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 4, dtype family float, shape (6, 5, 3, 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 numpy as np

torch.manual_seed(0)

# Create a 4D numpy array in HWIO format (Height, Width, Input channels, Output channels)
# Using distinct dimensions: H=3, W=4, I=5, O=6
conv_weights = np.random.randn(3, 4, 5, 6).astype(np.float32)

ARGS = {
    "conv_weights": conv_weights
}
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 Tensor, ndim 4, dtype family float, shape (6, 5, 3, 4) — shape (6, 5, 3, 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
topk Ran
ran; our contract draft was wrong, not the code
alextamkin/active-learning-pretrained-models/utils/train_test_common.py
code served (permissive licence) · get_code("228b4da98f348fbe")
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 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
import torch.nn as nn

torch.manual_seed(0)

# The function expects:
# - output: a tensor with a .topk() method (used as output.topk(max(ks), 1, True, True))
#   The topk call returns (values, indices) where indices has shape (batch_size, max(ks))
# - target: a tensor that can be viewed and compared with pred
# - ks: a tuple of integers

# Create small tensors for testing
batch_size = 2
num_classes = 4

# output should be logits/scores with shape (batch_size, num_classes)
output = torch.randn(batch_size, num_classes)

# target should have shape (batch_size,) with class indices
target = torch.tensor([0, 2], dtype=torch.long)

# ks is a tuple of k values to check
ks = (1, 2)

ARGS = {
    "output": output,
    "target": target,
    "ks": ks,
}
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 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
create_dat Not yet run
unverified (harvested, no recorded run)
alextamkin/active-learning-pretrained-models/utils/lbtoolbox.py
code served (permissive licence) · get_code("d38eeb648f68f197")
How it ran

no invocation recorded.

load_dat Not yet run
unverified (harvested, no recorded run)
alextamkin/active-learning-pretrained-models/utils/lbtoolbox.py
code served (permissive licence) · get_code("12109eac56e48b6b")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 8 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 2 the same month, 8 later) and 8 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
conv1x1 33 other papers
14 dated earlier, 1 the same month, 18 later · 3 only as an assembly brick
1912.11370 (2019-12)
Big Transfer (BiT): General Visual Representation Learning
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2102.04306 (2021-02)
TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation
and 30 more not named here
conv3x3 33 other papers
14 dated earlier, 1 the same month, 18 later · 3 only as an assembly brick
1912.11370 (2019-12)
Big Transfer (BiT): General Visual Representation Learning
2010.11929 (2020-10)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
2102.04306 (2021-02)
TransUNet: Transformers Make Strong Encoders for Medical Image Segmentation
and 30 more not named here
tf2th 17 other papers
6 dated earlier, 11 later
1912.11370 (2019-12)
Big Transfer (BiT): General Visual Representation Learning
2104.02610 (2021-04)
On the Robustness of Vision Transformers to Adversarial Examples
2106.01127 (2021-06)
Towards Robust Classification Model by Counterfactual and Invariant Data Generation
and 14 more not named here
create_dat 4 other papers
2 dated earlier, 2 later
2108.02833 (2021-08)
Elaborative Rehearsal for Zero-shot Action Recognition
2111.09805 (2021-11)
DICE: Leveraging Sparsification for Out-of-Distribution Detection
2205.09723 (2022-05)
Robust and Efficient Medical Imaging with Self-Supervision
and 1 more not named here
get_mixup 4 other papers
2 dated earlier, 2 later
2108.02833 (2021-08)
Elaborative Rehearsal for Zero-shot Action Recognition
2111.09805 (2021-11)
DICE: Leveraging Sparsification for Out-of-Distribution Detection
2205.09723 (2022-05)
Robust and Efficient Medical Imaging with Self-Supervision
and 1 more not named here
load_dat 4 other papers
2 dated earlier, 2 later
2108.02833 (2021-08)
Elaborative Rehearsal for Zero-shot Action Recognition
2111.09805 (2021-11)
DICE: Leveraging Sparsification for Out-of-Distribution Detection
2205.09723 (2022-05)
Robust and Efficient Medical Imaging with Self-Supervision
and 1 more not named here
setup_logger 2 other papers
1 dated earlier, 1 later
2108.02833 (2021-08)
Elaborative Rehearsal for Zero-shot Action Recognition
2205.09723 (2022-05)
Robust and Efficient Medical Imaging with Self-Supervision
topk 2 other papers
2 later
2206.09843 (2022-06)
Contextual Squeeze-and-Excitation for Efficient Few-Shot Image Classification
2212.10556 (2022-12)
Unleashing the Power of Visual Prompting At the Pixel Level

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

Models can fail in unpredictable ways during deployment due to task ambiguity, when multiple behaviors are consistent with the provided training data. An example is an object classifier trained on red squares and blue circles: when encountering blue squares, the intended behavior is undefined. We investigate whether pretrained models are better active learners, capable of disambiguating between the possible tasks a user may be trying to specify. Intriguingly, we find that better active learning is an emergent property of the pretraining process: pretrained models require up to 5 times fewer labels when using uncertainty-based active learning, while non-pretrained models see no or even negative benefit. We find these gains come from an ability to select examples with attributes that disambiguate the intended behavior, such as rare product categories or atypical backgrounds. These attributes are far more linearly separable in pretrained model's representation spaces vs non-pretrained models, suggesting a possible mechanism for this behavior.

For agents

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

get_harvested_code_for_paper("2204.08491")
get_code_for_paper("2204.08491")
have("2204.08491")

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

curl https://syntology.ai/api/ran/2204.08491.json

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

[![Syntology run record](https://syntology.ai/api/ran/2204.08491.svg)](https://syntology.ai/paper/2204.08491)

Connect an agent — have() is free.