SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1703.00767 · 2017

Attentive Recurrent Comparators

arXiv · PDF · Open in the Atlas

Code that ran

3 of 4 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 3 have no contract verdict; 0 failures of our instrument; 1 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/1703.00767.json · badge: /api/ran/1703.00767.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
gitabcworld/ConvArc community (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_accuracy_roc Ran
ran on a synthesized input
gitabcworld/ConvArc/baselineSiamese.py
code served (permissive licence) · get_code("51d88035f10c5d62")
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 numpy as np

# Create simple test data for binary classification
# labels: binary labels (0 or 1)
# predictions: prediction scores/probabilities

np.random.seed(0)

labels = np.array([0, 1, 1, 0, 1, 0, 1, 1])
predictions = np.array([0.1, 0.9, 0.8, 0.2, 0.7, 0.3, 0.85, 0.95])

ARGS = {
    "labels": labels,
    "predictions": predictions
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
context_train Ran
ran on a synthesized input
gitabcworld/ConvArc/context_train.py
code served (permissive licence) · get_code("5278d9683163b68d")
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 float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
import numpy as np
from datetime import datetime

torch.manual_seed(0)

# Create a minimal discriminator model
class MinimalDiscriminator(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create a minimal context_fn model
class MinimalContextFn(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create a minimal fcn model
class MinimalFCN(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create a minimal coAttn model
class MinimalCoAttn(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create a minimal options object
class Options:
    def __init__(self):
        self.cuda = False
        self.apply_wrn = True
        self.use_coAttn = True

# Create a minimal logger
class MinimalLogger:
    def log_value(self, name, value):
        pass

# Create a minimal epoch function that returns accuracies and losses
def minimal_epoch_fn(opt, loss_fn, discriminator, data_loader, model_fn, optimizer, fcn=None, coAttn=None):
    train_acc_epoch = [0.85, 0.87]
    train_loss_epoch = [0.15, 0.12]
    return train_acc_epoch, train_loss_epoch

# Create a minimal loss function
class MinimalLossFn(nn.Module):
    def __init__(self):
        super().__init__()
    
    def forward(self, x, y):
        return torch.tensor(0.1)

# Create a minimal optimizer
discriminator = MinimalDiscriminator()
context_fn = MinimalContextFn()
fcn = MinimalFCN()
coAttn = MinimalCoAttn()

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

# Create a minimal train loader (list of (x, y) tuples)
train_loader = [
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
]

# Create logger
logger = MinimalLogger()

# Create options
opt = Options()

# Create loss function
loss_fn = MinimalLossFn()

ARGS = {
    'epoch': 1,
    'epoch_fn': minimal_epoch_fn,
    'opt': opt,
    'train_loader': train_loader,
    'discriminator': discriminator,
    'context_fn': context_fn,
    'logger': logger,
    'optimizer': optimizer,
    'loss_fn': loss_fn,
    'fcn': fcn,
    'coAttn': coAttn,
}
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 1, dtype family float — 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
context_val Ran
ran on a synthesized input
gitabcworld/ConvArc/context_val.py
code served (permissive licence) · get_code("216eb13a53ae1b21")
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 float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 sys
import torch
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

# Create a minimal options object with required attributes
class Opt:
    def __init__(self):
        self.cuda = False
        self.apply_wrn = True
        self.use_coAttn = True
        self.val_num_batches = 2
        self.naive_full_save_path = "/tmp/model_checkpoint.pt"

# Create minimal models
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

class Discriminator(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

class ContextFn(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

class FCN(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

class CoAttn(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 2)
    
    def forward(self, x):
        return self.linear(x)

# Create a minimal logger
class Logger:
    def __init__(self):
        self.logs = {}
    
    def log_value(self, key, value):
        if key not in self.logs:
            self.logs[key] = []
        self.logs[key].append(value)

# Create a minimal epoch function
def epoch_fn(opt, loss_fn, discriminator, data_loader, model_fn, fcn=None, coAttn=None):
    val_acc = [0.8, 0.85]
    val_loss = [0.2, 0.15]
    return val_acc, val_loss

# Create a minimal data loader
val_loader = [
    (torch.randn(2, 4), torch.randint(0, 2, (2,))),
    (torch.randn(2, 4), torch.randint(0, 2, (2,)))
]

# Create loss function
loss_fn = nn.CrossEntropyLoss()

# Construct all arguments
ARGS = {
    "epoch": 1,
    "epoch_fn": epoch_fn,
    "opt": Opt(),
    "val_loader": val_loader,
    "discriminator": Discriminator(),
    "context_fn": ContextFn(),
    "logger": Logger(),
    "loss_fn": loss_fn,
    "fcn": FCN(),
    "coAttn": CoAttn()
}
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 1, dtype family float — 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
path_leaf Not yet run
unverified (harvested, no recorded run)
gitabcworld/ConvArc/baselineTriplets.py
code served (permissive licence) · get_code("0aae00daa68b5f3e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 4 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

Rapid learning requires flexible representations to quickly adopt to new evidence. We develop a novel class of models called Attentive Recurrent Comparators (ARCs) that form representations of objects by cycling through them and making observations. Using the representations extracted by ARCs, we develop a way of approximating a \textit{dynamic representation space} and use it for one-shot learning. In the task of one-shot classification on the Omniglot dataset, we achieve the state of the art performance with an error rate of 1.5\%. This represents the first super-human result achieved for this task with a generic model that uses only pixel information.

For agents

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

get_harvested_code_for_paper("1703.00767")
get_code_for_paper("1703.00767")
have("1703.00767")

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

curl https://syntology.ai/api/ran/1703.00767.json

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

[![Syntology run record](https://syntology.ai/api/ran/1703.00767.svg)](https://syntology.ai/paper/1703.00767)

Connect an agent — have() is free.