SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.04191 · ICLR · 2021

SI-Score: An image dataset for fine-grained analysis of robustness to object location, rotation and size

Josip Djolonga, Lucas Beyer, Alexander Kolesnikov, Xiaohua Zhai, Sylvain Gelly, Neil Houlsby, Jessica Yung, Rob Romijnders, Mario Lucic

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 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/2104.04191.json · badge: /api/ran/2104.04191.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
google-research/si-score canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
load_image Ran
ran, honoured the contract we drafted
google-research/si-score/dataset_generator.py
code served (permissive licence) · get_code("4b80482911c3de43")
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 PngImageFile. 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 io
from PIL import Image
import tempfile
import os

# Create a minimal valid image file
def create_test_image():
    # Create a simple PIL image
    img = Image.new('RGB', (10, 10), color='red')
    
    # Save to a temporary file
    temp_file = tempfile.NamedTemporaryFile(delete=False, suffix='.png')
    temp_path = temp_file.name
    temp_file.close()
    
    img.save(temp_path)
    return temp_path

# Create the test image and get its path
test_image_path = create_test_image()

ARGS = {
    'fname': test_image_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PngImageFile — a PngImageFile output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a PngImageFile output the recorded ndim 3 and dtype family int 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
resize_fg Ran
ran; our contract draft was wrong, not the code
google-research/si-score/dataset_generator.py
code served (permissive licence) · get_code("eb024909b3bf34dc")
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 Image. 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
import PIL
from PIL import Image

# Create small PIL Images for testing
# fg: foreground image - needs .copy(), .height, .width, .resize() methods
fg = Image.new('RGB', (100, 80))  # width=100, height=80

# bg: background image - needs .width, .height attributes
bg = Image.new('RGB', (400, 300))  # width=400, height=300

# fg_target_size: target size as percentage
fg_target_size = 25.0

ARGS = {
    'fg': fg,
    'bg': bg,
    'fg_target_size': fg_target_size,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Image — a Image output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Image output the recorded ndim 3 and dtype family int 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
validate_config Ran
ran; our contract draft was wrong, not the code
google-research/si-score/dataset_generator.py
code served (permissive licence) · get_code("6afcca506ca1c934")
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

ARGS = {
    'config': {
        'area': 100.5,
        'coord': [10.0, 20.0],
        'rotation': 45.0,
        'bg_resolution': [(1920, 1080), (1280, 720)]
    }
}
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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 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.

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

Before deploying machine learning models it is critical to assess their robustness. In the context of deep neural networks for image understanding, changing the object location, rotation and size may affect the predictions in non-trivial ways. In this work we perform a fine-grained analysis of robustness with respect to these factors of variation using SI-SCORE, a synthetic dataset. In particular, we investigate ResNets, Vision Transformers and CLIP, and identify interesting qualitative differences between these.

For agents

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

get_harvested_code_for_paper("2104.04191")
get_code_for_paper("2104.04191")
have("2104.04191")

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

curl https://syntology.ai/api/ran/2104.04191.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.04191.svg)](https://syntology.ai/paper/2104.04191)

Connect an agent — have() is free.