SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.00643 · ICML · 2021

Mandoline: Model Evaluation under Distribution Shift

Christopher Ré, Nimit Sohoni, Kayvon Fatahalian, Mayee Chen, Karan Goel, Fait Poms

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, 1 violated it and 0 have no contract verdict; 1 failure 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/2107.00643.json · badge: /api/ran/2107.00643.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
HazyResearch/mandoline canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
Phi Ran
ran, honoured the contract we drafted
HazyResearch/mandoline/mandoline.py
code served (permissive licence) · get_code("9244f2b360b12414")
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) and returned ndarray, ndim 2, dtype family int, shape (2, 6) 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

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

# Create a small n x d matrix of examples and slices
# n=2 examples, d=4 slices
D = np.random.choice([-1, 1], size=(2, 4))

# Create edge_list: k x 2 matrix of edge correlations
# Each row contains indices of two columns to correlate
edge_list = np.array([[0, 1], [1, 3]], dtype=np.int64)

ARGS = {
    "D": D,
    "edge_list": edge_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family int, shape (2, 6) — shape (2, 6) 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
log_partition_ratio Ran
ran, violated the contract we drafted
HazyResearch/mandoline/mandoline.py
code served (permissive licence) · get_code("e45d15f7a93aca51")
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) and returned float64, ndim 0, dtype family float, shape () 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 scipy.special

# Construct arguments for log_partition_ratio
# The function takes:
# - x: np.ndarray (a weight vector)
# - Phi_D_src: np.ndarray (a matrix of features)
# - n_src: int (number of source samples)

# x is a 1D weight vector
x = np.array([0.5, -0.3, 0.2, 0.1])

# Phi_D_src is a 2D matrix where rows are samples and columns are features
# It's used as Phi_D_src.dot(x), so it should have shape (n_samples, len(x))
# Using 5 samples and 4 features to match x dimension
Phi_D_src = np.random.randn(5, 4)

# n_src is the number of source samples
n_src = 5

ARGS = {
    "x": x,
    "Phi_D_src": Phi_D_src,
    "n_src": n_src,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float, shape () — shape () 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
mandoline Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
HazyResearch/mandoline/mandoline.py
code served (permissive licence) · get_code("347405051e7701e2")
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) and returned SimpleNamespace. 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
from functools import partial
from types import SimpleNamespace
import scipy.special
import scipy.optimize
import sklearn.metrics.pairwise as skmetrics

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

# Create small source distribution matrix (n_src x d)
# n_src=2 samples, d=4 features
D_src = np.random.choice([-1, 1], size=(2, 4))

# Create small target distribution matrix (n_tgt x d)
# n_tgt=3 samples, d=4 features
D_tgt = np.random.choice([-1, 1], size=(3, 4))

# Create edge list specifying correlations between feature pairs
# k=2 edges, each edge is a pair of column indices
edge_list = np.array([[0, 1], [1, 3]])

# sigma parameter for RBF kernel (None means standard KLIEP without kernel)
sigma = None

ARGS = {
    "D_src": D_src,
    "D_tgt": D_tgt,
    "edge_list": edge_list,
    "sigma": sigma,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, scipy, sklearn — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SimpleNamespace — a SimpleNamespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a SimpleNamespace 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

Machine learning models are often deployed in different settings than they were trained and validated on, posing a challenge to practitioners who wish to predict how well the deployed model will perform on a target distribution. If an unlabeled sample from the target distribution is available, along with a labeled sample from a possibly different source distribution, standard approaches such as importance weighting can be applied to estimate performance on the target. However, importance weighting struggles when the source and target distributions have non-overlapping support or are high-dimensional. Taking inspiration from fields such as epidemiology and polling, we develop Mandoline, a new evaluation framework that mitigates these issues. Our key insight is that practitioners may have prior knowledge about the ways in which the distribution shifts, which we can use to better guide the importance weighting procedure. Specifically, users write simple "slicing functions"-noisy, potentially correlated binary functions intended to capture possible axes of distribution shift-to compute reweighted performance estimates. We further describe a density ratio estimation framework for the slices and show how its estimation error scales with slice quality and dataset size. Empirical validation on NLP and vision tasks shows that Mandoline can estimate performance on the target distribution up to 3× more accurately compared to standard baselines.

For agents

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

get_harvested_code_for_paper("2107.00643")
get_code_for_paper("2107.00643")
have("2107.00643")

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

curl https://syntology.ai/api/ran/2107.00643.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.00643.svg)](https://syntology.ai/paper/2107.00643)

Connect an agent — have() is free.