SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2011.01837 · NeurIPS · 2020

The Gap on GAP: Tackling the Problem of Differing Data Distributions in Bias-Measuring Datasets

Thomas Lukasiewicz, Vid Kocijan, Oana-Maria Camburu

arXiv · PDF · Open in the Atlas

Code that ran

2 of 2 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 2 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/2011.01837.json · badge: /api/ran/2011.01837.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
vid-koci/weightingGAP canonical (classifier label)
classifier basis: census
2 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_scores Ran
ran on a synthesized input
vid-koci/weightingGAP/gap_scorer.py
code served (permissive licence) · get_code("db3dbbae0e8afc09")
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 json
import tempfile
import os

class Annotation:
    """Stand-in annotation class with required attributes."""
    def __init__(self, name_a_coref, name_b_coref, gender):
        self.name_a_coref = name_a_coref
        self.name_b_coref = name_b_coref
        self.gender = gender

# Create temporary weights file
weights_data = {
    'example1a': 0.8,
    'example1b': 0.9,
    'example2a': 0.7,
    'example2b': 0.85,
}

temp_dir = tempfile.mkdtemp()
weights_path = os.path.join(temp_dir, 'weights.json')
with open(weights_path, 'w') as f:
    json.dump(weights_data, f)

# Create gold annotations
gold_annotations = {
    'example1': Annotation(name_a_coref=True, name_b_coref=False, gender='M'),
    'example2': Annotation(name_a_coref=True, name_b_coref=True, gender='F'),
}

# Create system annotations
system_annotations = {
    'example1': Annotation(name_a_coref=True, name_b_coref=None, gender='M'),
    'example2': Annotation(name_a_coref=True, name_b_coref=True, gender='F'),
}

ARGS = {
    'gold_annotations': gold_annotations,
    'system_annotations': system_annotations,
    'weights_path': weights_path,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
compare Ran
ran on a synthesized input
vid-koci/weightingGAP/compute_weights.py
code served (permissive licence) · get_code("b224cb4b769df69d")
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 bool, 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 torch

torch.manual_seed(0)

# span_a and span_b are tuples/lists with at least 2 elements
# The function accesses span_a[0], span_a[1], span_b[0], span_b[1]
# and performs comparison operations on them
span_a = (1, 5)
span_b = (3, 8)

ARGS = {
    "span_a": span_a,
    "span_b": span_b
}
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 bool, 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

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 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 2 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

Diagnostic datasets that can detect biased models are an important prerequisite for bias reduction within natural language processing. However, undesired patterns in the collected data can make such tests incorrect. For example, if the feminine subset of a gender-bias-measuring coreference resolution dataset contains sentences with a longer average distance between the pronoun and the correct candidate, an RNNbased model may perform worse on this subset due to longterm dependencies. In this work, we introduce a theoretically grounded method for weighting test samples to cope with such patterns in the test data. We demonstrate the method on the GAP dataset for coreference resolution. We annotate GAP with spans of all personal names and show that examples in the female subset contain more personal names and a longer distance between pronouns and their referents, potentially affecting the bias score in an undesired way. Using our weighting method, we find the set of weights on the test instances that should be used for coping with these correlations, and we re-evaluate 16 recently released coreference models. 1

For agents

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

get_harvested_code_for_paper("2011.01837")
get_code_for_paper("2011.01837")
have("2011.01837")

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

curl https://syntology.ai/api/ran/2011.01837.json

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

[![Syntology run record](https://syntology.ai/api/ran/2011.01837.svg)](https://syntology.ai/paper/2011.01837)

Connect an agent — have() is free.