SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.09618 · ICML · 2023

Emergent Asymmetry of Precision and Recall for Measuring Fidelity and Diversity of Generative Models in High Dimensions

Wael Abdalmageed, Mahyar Khayatkhoei

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; 0 honoured the contract we drafted, 0 violated it and 3 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/2306.09618.json · badge: /api/ran/2306.09618.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
mahyarkoy/emergent_asymmetry_pr canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
add_color_borders Ran
ran on a synthesized input
mahyarkoy/emergent_asymmetry_pr/manifmetric/dataset.py
code served (permissive licence) · get_code("caaf0d5d60f493d0")
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 ndarray, ndim 4, dtype family float, shape (2, 8, 7, 3) 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 numpy as np
import matplotlib

# Set up seed for reproducibility
np.random.seed(0)

# Create small test images: batch_size=2, height=4, width=3, channels=1
# Values in [0, 1]
imgs = np.random.rand(2, 4, 3, 1).astype(np.float32)

# Create labels: batch_size=2, with integer values
labels = np.array([0, 2], dtype=np.int32)

# max_label parameter
max_label = 3

# color_map parameter - use a valid matplotlib colormap name
color_map = "viridis"

# color_set parameter - shape (max_label+1, 3) with RGB values
color_set = np.random.rand(4, 3).astype(np.float32)

# Border sizes
fh = 2
fw = 2

ARGS = {
    "imgs": imgs,
    "labels": labels,
    "max_label": max_label,
    "color_map": color_map,
    "color_set": color_set,
    "fh": fh,
    "fw": fw
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
matplotlib, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 4, dtype family float, shape (2, 8, 7, 3) — shape (2, 8, 7, 3) 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
block_draw Ran
ran on a synthesized input
mahyarkoy/emergent_asymmetry_pr/manifmetric/dataset.py
code served (permissive licence) · get_code("dd7206ffdf8e4490")
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 ndarray, ndim 3, dtype family int, shape (8, 3, 3) 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 numpy as np
import tempfile

# Create simple test images: (batch=2, height=4, width=3, channels=3)
# Values in [0, 1]
imgs = np.random.rand(2, 4, 3, 3).astype(np.float32)

# Create a temporary path for saving
temp_file = tempfile.NamedTemporaryFile(suffix='.png', delete=False)
save_path = temp_file.name
temp_file.close()

ARGS = {
    'imgs': imgs,
    'path': save_path,
    'border': False,
    'rows': None,
    'cols': None,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, matplotlib, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (8, 3, 3) — shape (8, 3, 3) 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
read_image Ran
ran on a synthesized input
mahyarkoy/emergent_asymmetry_pr/manifmetric/dataset.py
code served (permissive licence) · get_code("4c65431b46d9b300")
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 ndarray, ndim 3, dtype family float, shape (64, 64, 3) 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 numpy as np
from PIL import Image
import tempfile
import os

# Create a temporary directory for test images
temp_dir = tempfile.mkdtemp()

# Create a simple test image (RGB, 100x100 pixels)
def create_test_image(path, width=100, height=100):
    img = Image.new('RGB', (width, height), color=(128, 64, 192))
    img.save(path)
    return path

# Create test image file
test_image_path = os.path.join(temp_dir, "test_image.png")
create_test_image(test_image_path, width=100, height=80)

# Define arguments for the function
ARGS = {
    "img_path": test_image_path,
    "img_size": 64,
    "sqcrop": False,
    "bbox": None,
    "verbose": False,
    "center_crop": None,
    "crop_size": None,
}
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 ndarray, ndim 3, dtype family float, shape (64, 64, 3) — shape (64, 64, 3) 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 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

Precision and Recall are two prominent metrics of generative performance, which were proposed to separately measure the fidelity and diversity of generative models. Given their central role in comparing and improving generative models, understanding their limitations are crucially important. To that end, in this work, we identify a critical flaw in the common approximation of these metrics using k-nearest-neighbors, namely, that the very interpretations of fidelity and diversity that are assigned to Precision and Recall can fail in high dimensions, resulting in very misleading conclusions. Specifically, we empirically and theoretically show that as the number of dimensions grows, two model distributions with supports at equal point-wise distance from the support of the real distribution, can have vastly different Precision and Recall regardless of their respective distributions, hence an emergent asymmetry in high dimensions. Based on our theoretical insights, we then provide simple yet effective modifications to these metrics to construct symmetric metrics regardless of the number of dimensions. Finally, we provide experiments on real-world datasets to illustrate that the identified flaw is not merely a pathological case, and that our proposed metrics are effective in alleviating its impact.

For agents

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

get_harvested_code_for_paper("2306.09618")
get_code_for_paper("2306.09618")
have("2306.09618")

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

curl https://syntology.ai/api/ran/2306.09618.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.09618.svg)](https://syntology.ai/paper/2306.09618)

Connect an agent — have() is free.