SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.14263 · 2021

Batch Active Learning at Scale

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 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 8 have no contract verdict; 0 failures of our instrument; 0 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/2107.14263.json · badge: /api/ran/2107.14263.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
airi-institute/al_toolbox community (classifier label)
classifier basis not recorded on the link
8 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_grad Ran
ran on a synthesized input
airi-institute/al_toolbox/acleto/post_processing/calculate_tracin_score.py
code served (permissive licence) · get_code("2cc02afb97254ce3")
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 (10,) 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 minimal model that:
# 1. Has parameters (so next(model.parameters()).device works)
# 2. Has a forward method that accepts keyword arguments
# 3. Returns a dict with 'loss' key
# 4. The loss is a scalar tensor that can be backpropagated
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x, y=None):
        # Accept x and y as keyword arguments
        output = self.linear(x)
        loss = output.sum()
        return {'loss': loss}

model = MinimalModel()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)

# Create point as a dict with tensor values
# The function expects a dict with keys that match forward's parameter names
point = {
    'x': torch.randn(2, 4),
    'y': torch.randn(2, 2)
}

ARGS = {
    'model': model,
    'optimizer': optimizer,
    'point': point
}
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 (10,) — shape (10,) 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
calculate_grad_flair Ran
ran on a synthesized input
airi-institute/al_toolbox/acleto/post_processing/calculate_tracin_score.py
code served (permissive licence) · get_code("576acc9de24783fe")
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 (15,) 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 SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward_loss(self, point):
        # Returns a tuple where first element is the loss tensor
        output = self.linear(point)
        loss = output.sum()
        return (loss,)

model = SimpleModel()
optimizer = torch.optim.SGD(model.parameters(), lr=0.01)
point = torch.randn(2, 4)

ARGS = {
    "model": model,
    "optimizer": optimizer,
    "point": point
}
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 (15,) — shape (15,) 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
convert_y_to_dict_format Ran
ran on a synthesized input
airi-institute/al_toolbox/jupyterlab_demo/utils_data.py
code served (permissive licence) · get_code("fd43f053d6bc8e97")
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 numpy as np

# Analyze the function:
# - X: iterable of sequences, each sequence is iterable of words (strings)
# - y: iterable of sequences, each sequence is iterable of tags (strings with format like 'B-PER', 'I-PER', 'O')
# - Tags are split by '-' and first part checked for 'O', 'B', 'I'
# - Words are accessed by index and have len()
# - Tags are split by '-' and indexed with [0] and [1]

# Create minimal valid inputs
X = [
    ['John', 'Smith', 'works'],
    ['Apple', 'Inc']
]

y = [
    ['B-PER', 'I-PER', 'O'],
    ['B-ORG', 'I-ORG']
]

ARGS = {
    'X': X,
    'y': y
}
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 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_custom_path Ran
ran on a synthesized input
airi-institute/al_toolbox/jupyterlab_demo/demo_system.py
code served (permissive licence) · get_code("7084c56588c1e6af")
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 str. 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
import tempfile

def create_custom_path(path):
    if not os.path.isdir(path):
        os.makedirs(path, exist_ok=True)
    return path

# Create a temporary directory to use as the parent
temp_dir = tempfile.gettempdir()
test_path = os.path.join(temp_dir, "test_custom_path_dir")

ARGS = {
    "path": test_path
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
create_helper Ran
ran on a synthesized input
airi-institute/al_toolbox/jupyterlab_demo/utils_data.py
code served (permissive licence) · get_code("316cb1f89e1cee44")
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 pandas as pd

# The function expects X_train to be an iterable (list) where each element
# is a tokenized_text (iterable of words/strings)
X_train = [
    ['hello', 'world', 'foo'],
    ['bar', 'baz'],
]

ARGS = {
    'X_train': X_train,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — 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
format_entities Ran
ran on a synthesized input
airi-institute/al_toolbox/jupyterlab_demo/utils_data.py
code served (permissive licence) · get_code("3117d3ae4991e45d")
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 Token:
    def __init__(self, begin, end, text):
        self.begin = begin
        self.end = end
        self.text = text

sentence = [
    Token(begin=0, end=5, text="Hello"),
    Token(begin=6, end=10, text="world"),
    Token(begin=11, end=13, text="is"),
    Token(begin=14, end=18, text="nice"),
]

tags = ["B-GREETING", "O", "B-ADJ", "O"]

ARGS = {
    "sentence": sentence,
    "tags": tags,
}
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
instance_to_sentence Ran
ran on a synthesized input
airi-institute/al_toolbox/domain_adaptation/hf_dataset_to_sent.py
code served (permissive licence) · get_code("e95ff7d47060fbd2")
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 str. 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)

# Analyzing the function:
# - string: needs to be a string (concatenated with instance)
# - instance: can be either:
#   1. A string (direct concatenation works)
#   2. An iterable (joined with spaces via ' '.join(instance))

# Test case 1: instance is a string
string_value = "Initial text"
instance_value = "some text to append"

ARGS = {
    "string": string_value,
    "instance": instance_value
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
point_to_device Ran
ran on a synthesized input
airi-institute/al_toolbox/acleto/post_processing/calculate_tracin_score.py
code served (permissive licence) · get_code("e17c116467304c37")
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 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# Test case 1: point is a torch.Tensor
tensor_point = torch.randn(2, 4)

# Test case 2: point is a list of tensors
list_point = [torch.randn(2, 4), torch.randn(2, 4)]

# Test case 3: point is a dict where values are lists (will be stacked)
dict_list_point = {
    'key1': [torch.randn(2, 4), torch.randn(2, 4)],
    'key2': [torch.randn(2, 4), torch.randn(2, 4)]
}

# Test case 4: point is a dict where values are tensors
dict_tensor_point = {
    'key1': torch.randn(2, 4),
    'key2': torch.randn(2, 4)
}

device = 'cpu'

# Use one of the test cases for ARGS
ARGS = {
    'point': tensor_point,
    'device': device
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 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.

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

The ability to train complex and highly effective models often requires an abundance of training data, which can easily become a bottleneck in cost, time, and computational resources. Batch active learning, which adaptively issues batched queries to a labeling oracle, is a common approach for addressing this problem. The practical benefits of batch sampling come with the downside of less adaptivity and the risk of sampling redundant examples within a batch -- a risk that grows with the batch size. In this work, we analyze an efficient active learning algorithm, which focuses on the large batch setting. In particular, we show that our sampling method, which combines notions of uncertainty and diversity, easily scales to batch sizes (100K-1M) several orders of magnitude larger than used in previous studies and provides significant improvements in model training efficiency compared to recent baselines. Finally, we provide an initial theoretical analysis, proving label complexity guarantees for a related sampling method, which we show is approximately equivalent to our sampling method in specific settings.

For agents

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

get_harvested_code_for_paper("2107.14263")
get_code_for_paper("2107.14263")
have("2107.14263")

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

curl https://syntology.ai/api/ran/2107.14263.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.14263.svg)](https://syntology.ai/paper/2107.14263)

Connect an agent — have() is free.