SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2104.02215 · ICCV · 2021

When Pigs Fly: Contextual Reasoning in Synthetic and Natural Scenes

Gabriel Kreiman, Mengmi Zhang, Spandan Madan, Philipp Bomatter, Dimitar Karev, Claire Tseng

arXiv · PDF · Open in the Atlas

Code that ran

3 of 6 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 0 have no contract verdict; 3 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/2104.02215.json · badge: /api/ran/2104.02215.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
kreimanlab/WhenPigsFlyContext canonical (classifier label)
classifier basis: census
3 ran · 3 no recorded run
FunctionStatusWhere it lives
build_uncertainty_gate Ran
ran; our contract draft was wrong, not the code
kreimanlab/WhenPigsFlyContext/core/model.py
code served (permissive licence) · get_code("721d8ac52576ca7f")
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 UncertaintyGate. 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 = {
    'num_encoder_features': 64,
    'num_classes': 10,
    'gate_type': 'entropy'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchvision — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type UncertaintyGate — a UncertaintyGate output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a UncertaintyGate 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
cuboid Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kreimanlab/WhenPigsFlyContext/unity/display_graph.py
code served (permissive licence) · get_code("b3ca8a1b900d75f3")
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 Poly3DCollection. 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 matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D

# Create a 3D plot
fig = plt.figure()
ax = fig.add_subplot(111, projection='3d')

# Define arguments for the cuboid function
# coords: [x, y, z] center coordinates
coords = [0.0, 1.0, 2.0]

# sides: [width, height, depth] dimensions
sides = [2.0, 3.0, 4.0]

ARGS = {
    'ax': ax,
    'coords': coords,
    'sides': sides
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, mpl_toolkits — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Poly3DCollection — a Poly3DCollection output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Poly3DCollection 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
display_grid_img Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
kreimanlab/WhenPigsFlyContext/unity/MMutils.py
code served (permissive licence) · get_code("2e2a843780e52b94")
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

# Create a list of sample images as numpy arrays
# Each image is (height, width, 3) with uint8 values (RGB format)
image1 = np.random.randint(0, 256, (4, 6, 3), dtype=np.uint8)
image2 = np.random.randint(0, 256, (4, 6, 3), dtype=np.uint8)
image3 = np.random.randint(0, 256, (4, 6, 3), dtype=np.uint8)

images_old = [image1, image2, image3]

ARGS = {
    'images_old': images_old,
    'nrows': 2
}
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
display_scene_modalities Not yet run
unverified (harvested, no recorded run)
kreimanlab/WhenPigsFlyContext/unity/MMutils.py
code served (permissive licence) · get_code("799ad751135e3b25")
How it ran

no invocation recorded.

find_nodes Not yet run
unverified (harvested, no recorded run)
kreimanlab/WhenPigsFlyContext/unity/MMutils.py
code served (permissive licence) · get_code("8e3413a872edc331")
How it ran

no invocation recorded.

get_xyz_mouse_click Not yet run
unverified (harvested, no recorded run)
kreimanlab/WhenPigsFlyContext/unity/display_graph.py
code served (permissive licence) · get_code("fc264d88eed3353e")
How it ran

no invocation recorded.

Also present in other papers' code

Of 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 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
cuboid 1 other paper
1 later
2202.01771 (2022-02)
Pre-Trained Language Models for Interactive Decision-Making
find_nodes 1 other paper
1 later
2202.01771 (2022-02)
Pre-Trained Language Models for Interactive Decision-Making
get_xyz_mouse_click 1 other paper
1 later
2202.01771 (2022-02)
Pre-Trained Language Models for Interactive Decision-Making

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

Images under normal context and out-of-context conditions were generated in the VirtualHome environment [27] using the Unity 3D simulation engine [19]. The same target object (a mug, red bounding box) is shown in different context conditions: normal context (a, b) and out-of-context conditions including gravity ((c), target object is floating in the air), change in object co-occurrence statistics (d), combination of both gravity and object co-occurrence statistics (e), enlarged object size (f), and no context with uniform grey pixels as background (g).

For agents

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

get_harvested_code_for_paper("2104.02215")
get_code_for_paper("2104.02215")
have("2104.02215")

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

curl https://syntology.ai/api/ran/2104.02215.json

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

[![Syntology run record](https://syntology.ai/api/ran/2104.02215.svg)](https://syntology.ai/paper/2104.02215)

Connect an agent — have() is free.