SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2210.05769 · 2022

Vote'n'Rank: Revision of Benchmarking with Social Choice Theory

arXiv · PDF · Open in the Atlas

Code that ran

6 of 17 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 6 have no contract verdict; 0 failures of our instrument; 11 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/2210.05769.json · badge: /api/ran/2210.05769.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
pragmaticslab/vote_and_rank canonical (classifier label)
classifier basis not recorded on the link
6 ran · 11 no recorded run
FunctionStatusWhere it lives
get_res_df Ran
ran on a synthesized input
pragmaticslab/vote_and_rank/votenrank/stability_exp.py
code served (permissive licence) · get_code("24983cbbf10c380a")
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 DataFrame. 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 pandas as pd
import numpy as np

# Construct valid arguments for get_res_df
exp_range = [0.1, 0.2, 0.3, 0.4, 0.5]

exp_res = {
    'method_a': [0.85, 0.87, 0.89, 0.91, 0.92],
    'method_b': [0.80, 0.82, 0.84, 0.86, 0.88],
    'method_c': [0.88, 0.90, 0.91, 0.93, 0.94]
}

ARGS = {
    'exp_range': exp_range,
    'exp_res': exp_res
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
mean_ranking Ran
ran on a synthesized input
pragmaticslab/vote_and_rank/votenrank/leaderboard/_rules.py
code served (permissive licence) · get_code("2753ffd283ace2c7")
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 Series. 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 pandas as pd
import numpy as np
from scipy.stats import gmean

# Create a minimal class that has the attributes/methods used by mean_ranking
class MockRanker:
    def __init__(self):
        # Create a small table with some numerical data
        # 4 rows (items to rank), 3 columns (metrics)
        self.table = pd.DataFrame({
            'metric1': [10.0, 20.0, 15.0, 25.0],
            'metric2': [12.0, 18.0, 14.0, 22.0],
            'metric3': [11.0, 19.0, 13.0, 23.0]
        }, index=['item1', 'item2', 'item3', 'item4'])
        
        # Weights for each metric
        self.weights = pd.Series([0.5, 0.3, 0.2])
        
        # Flag for partial data (whether there are NaN values)
        self.is_partial = False

# Instantiate the mock ranker
mock_ranker = MockRanker()

# Define ARGS with all parameters to mean_ranking
ARGS = {
    'self': mock_ranker,
    'mean_type': 'arithmetic'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Series — a Series output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Series output the recorded ndim 1 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
naive_masking_score Ran
ran on a synthesized input
pragmaticslab/vote_and_rank/votenrank/fairness_computation.py
code served (permissive licence) · get_code("61ecad216e42085c")
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 float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is not recorded. If your copy returns the same type and ndim 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
import numpy as np

torch.manual_seed(0)

class MockTokenizer:
    def __init__(self):
        self.mask_token_id = 103
    
    def encode(self, sentence, return_tensors=None):
        # Return a small tensor of token IDs
        # Shape: (1, seq_len) where seq_len = 5
        tokens = torch.tensor([[101, 2054, 2003, 2009, 102]])
        return tokens

class MockModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 1)
    
    def forward(self, input_ids, labels=None):
        # input_ids shape: (batch_size, seq_len)
        # Return an object with .loss attribute
        batch_size = input_ids.size(0)
        # Compute a simple loss
        loss_value = torch.tensor(0.5)
        
        class Output:
            def __init__(self, loss):
                self.loss = loss
        
        return Output(loss_value)

model = MockModel()
tokenizer = MockTokenizer()
sentence = "What is this?"

ARGS = {
    "model": model,
    "tokenizer": tokenizer,
    "sentence": sentence
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — 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 — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
naive_t5_score Ran
ran on a synthesized input
pragmaticslab/vote_and_rank/votenrank/fairness_computation.py
code served (permissive licence) · get_code("335d437c607e19d1")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
import torch.nn as nn

torch.manual_seed(0)

class MockTokenizer:
    def encode(self, text, return_tensors=None):
        # Return a small token sequence
        # For '<extra_id_0>' return a specific id, for '<extra_id_1>' return another
        if text == '<extra_id_0>':
            tokens = [32099]
        elif text == '<extra_id_1>':
            tokens = [32098]
        else:
            # For actual sentences, return a small sequence of token ids
            tokens = [101, 2054, 2003, 102]  # [CLS] what is [SEP]
        
        if return_tensors == 'pt':
            return torch.tensor([tokens])
        return tokens

class MockT5Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.embedding = nn.Embedding(32100, 8)
        self.linear = nn.Linear(8, 32100)
    
    def forward(self, input_ids, labels=None):
        # Minimal forward pass that accepts input_ids and labels
        embeddings = self.embedding(input_ids)
        logits = self.linear(embeddings)
        
        if labels is not None:
            # Compute a dummy loss
            loss_fn = nn.CrossEntropyLoss()
            loss = loss_fn(logits.view(-1, 32100), labels.view(-1))
        else:
            loss = None
        
        return type('Output', (), {'loss': loss})()

model = MockT5Model()
tokenizer = MockTokenizer()
sentence = "what is this"

ARGS = {
    'model': model,
    'tokenizer': tokenizer,
    'sentence': sentence
}
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 float, 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
preprocess_value Ran
ran on a synthesized input
pragmaticslab/vote_and_rank/votenrank/data_processing.py
code served (permissive licence) · get_code("40e3b0bc93d87e29")
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 DataFrame. 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 pandas as pd

def preprocess_value(value):
    value = value.set_index('Model').drop(columns=['Mean-Rank', 'Meta-Ave'])
    value = value.replace({'-': np.nan})
    value = value.astype(float)
    value.index = ['Human', 'craig.starr', 'DuKG', 'HERO 1', 'HERO 2', 'HERO 3', 'HERO 4']
    return value

# Create a sample DataFrame that matches what the function expects
data = {
    'Model': ['Human', 'craig.starr', 'DuKG', 'HERO 1', 'HERO 2', 'HERO 3', 'HERO 4'],
    'Mean-Rank': [1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0],
    'Meta-Ave': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7],
    'Metric1': [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3],
    'Metric2': [0.5, 0.6, '-', 0.8, 0.9, '-', 0.4],
    'Metric3': [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7]
}

df = pd.DataFrame(data)

ARGS = {
    'value': df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
ranking2top Ran
ran on a synthesized input
pragmaticslab/vote_and_rank/votenrank/utils.py
code served (permissive licence) · get_code("708a0312fba2e239")
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 list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 pandas as pd
import numpy as np

# Create a pandas Series that acts as a ranking
# The function accesses .max(), indexes with ==, and calls .index.tolist()
ranking_data = pd.Series([0.1, 0.5, 0.5, 0.2], index=['a', 'b', 'c', 'd'])

ARGS = {
    'ranking': ranking_data
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
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
agreement_rate Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/utils.py
code served (permissive licence) · get_code("62d6f8c97a21042f")
How it ran

no invocation recorded.

compute_iia Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/iia_exp.py
code served (permissive licence) · get_code("63a3462e0df18dbf")
How it ran

no invocation recorded.

compute_iia_for_fixed_models Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/iia_exp.py
code served (permissive licence) · get_code("f391b7410d101195")
How it ran

no invocation recorded.

fine_sorted_ranking Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/iia_exp.py
code served (permissive licence) · get_code("99837b3c6b7db771")
How it ran

no invocation recorded.

kendall_tau Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/utils.py
code served (permissive licence) · get_code("c7b9ab94d004f31d")
How it ran

no invocation recorded.

mean_election Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/leaderboard/_rules.py
code served (permissive licence) · get_code("0b38e82dfe604c4a")
How it ran

no invocation recorded.

naive_gpt2_score Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/fairness_computation.py
code served (permissive licence) · get_code("fb36e5dc20e0a615")
How it ran

no invocation recorded.

plurality_ranking Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/leaderboard/_rules.py
code served (permissive licence) · get_code("259e4bac30e16d32")
How it ran

no invocation recorded.

preprocess_glue Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/data_processing.py
code served (permissive licence) · get_code("ad5284fde4881978")
How it ran

no invocation recorded.

preprocess_sglue Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/data_processing.py
code served (permissive licence) · get_code("dbcd5b79074ddddd")
How it ran

no invocation recorded.

spearman_exp Not yet run
unverified (harvested, no recorded run)
pragmaticslab/vote_and_rank/votenrank/stability_exp.py
code served (permissive licence) · get_code("a6b079a0864a73e8")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 17 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 17 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

The development of state-of-the-art systems in different applied areas of machine learning (ML) is driven by benchmarks, which have shaped the paradigm of evaluating generalisation capabilities from multiple perspectives. Although the paradigm is shifting towards more fine-grained evaluation across diverse tasks, the delicate question of how to aggregate the performances has received particular interest in the community. In general, benchmarks follow the unspoken utilitarian principles, where the systems are ranked based on their mean average score over task-specific metrics. Such aggregation procedure has been viewed as a sub-optimal evaluation protocol, which may have created the illusion of progress. This paper proposes Vote'n'Rank, a framework for ranking systems in multi-task benchmarks under the principles of the social choice theory. We demonstrate that our approach can be efficiently utilised to draw new insights on benchmarking in several ML sub-fields and identify the best-performing systems in research and development case studies. The Vote'n'Rank's procedures are more robust than the mean average while being able to handle missing performance scores and determine conditions under which the system becomes the winner.

For agents

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

get_harvested_code_for_paper("2210.05769")
get_code_for_paper("2210.05769")
have("2210.05769")

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

curl https://syntology.ai/api/ran/2210.05769.json

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

[![Syntology run record](https://syntology.ai/api/ran/2210.05769.svg)](https://syntology.ai/paper/2210.05769)

Connect an agent — have() is free.