SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.06751 · ACL · 2024

The Unreasonable Effectiveness of Easy Training Data for Hard Tasks

Peter Clark, Sarah Wiegreffe, Peter Hase, Mohit Bansal

arXiv · PDF · Open in the Atlas

Code that ran

12 of 13 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; 2 honoured the contract we drafted, 0 violated it and 1 has no contract verdict; 9 failures of our instrument; 1 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/2401.06751.json · badge: /api/ran/2401.06751.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
allenai/easy-to-hard-generalization canonical (classifier label)
classifier basis: census
12 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_mc_loss Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
allenai/easy-to-hard-generalization/utils/LM_utils.py
code served (permissive licence) · get_code("e46136d7a352b173")
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 tuple. 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

torch.manual_seed(0)

# Build arguments for compute_mc_loss(pred_probs, answer_idx)
# pred_probs: tensor of shape (batch_size, num_classes) with probabilities
# answer_idx: tensor of shape (batch_size,) with class indices

batch_size = 2
num_classes = 4

# pred_probs: probability distribution over classes for each sample
pred_probs = torch.softmax(torch.randn(batch_size, num_classes), dim=1)

# answer_idx: ground truth class indices
answer_idx = torch.tensor([1, 2])

ARGS = {
    "pred_probs": pred_probs,
    "answer_idx": answer_idx,
}
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 tuple — a tuple output has no array shape of its own; none is 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
force_not_dimensionless Ran
ran, honoured the contract we drafted
allenai/easy-to-hard-generalization/utils/metrics.py
code served (permissive licence) · get_code("48c9a8969754d890")
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 Tensor, ndim 1, dtype family float, shape (1,) 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)

# The function checks if data is a torch.Tensor and if it has 0 dimensions (scalar)
# If so, it reshapes it to have 1 dimension
# We need to test with a 0-dimensional tensor (scalar tensor)

ARGS = {
    "data": torch.tensor(5.0)  # 0-dimensional tensor
}
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 Tensor, ndim 1, dtype family float, shape (1,) — shape (1,) 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
gather_item_level_stats_df Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/utils/utils.py
code served (permissive licence) · get_code("0a5e30383bb5d45e")
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

# Create sample DataFrames with required structure
def create_sample_item_df(n_items=5, seed=0):
    np.random.seed(seed)
    df = pd.DataFrame({
        'accuracy': np.random.rand(n_items),
        'difficulty': np.random.choice(['easy', 'medium', 'hard'], n_items),
        'difficulty_level': np.random.randint(1, 4, n_items)
    })
    return df

# Build the nested dictionary structure required by the function
# id_no_boot_to_item_accs[dataname][prompt_id][n_train][boot_idx]
id_no_boot_to_item_accs = {
    'dataset_1': {
        'prompt_001': {
            10: [
                create_sample_item_df(n_items=5, seed=0),
                create_sample_item_df(n_items=5, seed=1),
                create_sample_item_df(n_items=5, seed=2),
            ]
        }
    }
}

ARGS = {
    'id_no_boot_to_item_accs': id_no_boot_to_item_accs,
    'n_train': 10,
    'dataname': 'dataset_1',
    'prompt_id': 'prompt_001',
    'hardness_col_name': 'difficulty'
}
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
get_all_possible_hardness_col_names Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/utils/utils.py
code served (permissive licence) · get_code("c4d81b835f803a9a")
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 torch
import numpy as np

# The function get_all_possible_hardness_col_names takes a single parameter: model_name
# It's a string that gets used in string formatting within get_hardness_col_names

ARGS = {
    "model_name": "bert-base"
}
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 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
get_base_command Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/run_jobs.py
code served (permissive licence) · get_code("0fbb4515a3f97b76")
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 str. 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 numpy as np

class Args:
    def __init__(self, gpu):
        self.gpu = gpu

torch.manual_seed(0)

ARGS = {
    'args': Args(gpu=0)
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
get_config_from_args Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/run_jobs.py
code served (permissive licence) · get_code("872f5fe645cf2260")
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 torch
import torch.nn as nn

torch.manual_seed(0)

class MockArgs:
    def __init__(self):
        self.learning_rate = 0.001
        self.batch_size = 32
        self.num_epochs = 10
        self.model_type = 'transformer'
        # Note: intentionally not setting all_attributes to test the 'NA' case

attrs_list = [
    'learning_rate',
    'batch_size',
    'num_epochs',
    'model_type',
    'missing_attribute',
    'another_missing'
]

args = MockArgs()

ARGS = {
    'args': args,
    'attrs': attrs_list
}
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 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
get_hardness_col_names Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/utils/utils.py
code served (permissive licence) · get_code("99af85a870a51904")
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 torch

torch.manual_seed(0)

ARGS = {
    'model_name': 'gpt2',
    'normed': False
}
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 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
grid_bootstrap Ran
ran on a synthesized input
allenai/easy-to-hard-generalization/utils/metrics.py
code served (permissive licence) · get_code("0a2e48585058c30c")
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 numpy as np

def p_value(betas):
    abs_mean_beta = np.abs(np.mean(betas))
    centered_betas = betas - np.mean(betas)
    outside_prop = np.mean(centered_betas < -abs_mean_beta) + np.mean(centered_betas > abs_mean_beta)
    return outside_prop

# Create a small test ndarray with shape (5 rows, 3 columns)
# Mix of numeric values and some NaN values
np.random.seed(42)
test_ndarray = np.random.randn(5, 3).astype(np.float64)
# Introduce a few NaN values but not entire rows
test_ndarray[0, 1] = np.nan
test_ndarray[2, 0] = np.nan

# Define a simple summary function that computes mean (ignoring NaN)
def summary_func(data):
    return np.nanmean(data)

ARGS = {
    'ndarray': test_ndarray,
    'summary_function': summary_func,
    'boot_times': 100
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
load_optimizer_and_scheduler Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/utils/modeling_utils.py
code served (permissive licence) · get_code("7ed61b42e40ce4dd")
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 tuple, 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 torch
import torch.nn as nn
from torch.optim import SGD, AdamW
from transformers import get_scheduler

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 8)
        self.bias_layer = nn.Linear(8, 2)
        self.layer_norm = nn.LayerNorm(8)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.layer_norm(x)
        x = self.bias_layer(x)
        return x

class ArgsConfig:
    def __init__(self, weight_decay=0.01, lr=1e-4, optimizer='adamw', lr_decay='linear'):
        self.weight_decay = weight_decay
        self.lr = lr
        self.optimizer = optimizer
        self.lr_decay = lr_decay

model = SimpleModel()
args = ArgsConfig(weight_decay=0.01, lr=1e-4, optimizer='adamw', lr_decay='linear')
num_training_steps = 1000

ARGS = {
    'args': args,
    'model': model,
    'num_training_steps': num_training_steps
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, transformers — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple 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
p_value Ran
ran, honoured the contract we drafted
allenai/easy-to-hard-generalization/utils/metrics.py
code served (permissive licence) · get_code("ecdceb0edcf86391")
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. 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 numpy as np

# The function p_value takes betas as input
# It expects a numpy array of beta values
# The function computes statistics on the betas using numpy operations

# Create a small test array of beta values
betas = np.array([0.5, 1.2, -0.3, 0.8, -0.1, 0.9])

ARGS = {
    "betas": betas
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — 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
renormalize_mc_pred_probs Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
allenai/easy-to-hard-generalization/utils/LM_utils.py
code served (permissive licence) · get_code("7bd4422fc24bf4a5")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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)

# Create a small tensor of shape (2, 4) with positive values
# This represents predicted probabilities for a batch of 2 samples with 4 classes
pred_probs = torch.rand(2, 4) + 0.1  # Add 0.1 to ensure all values are positive

ARGS = {
    "pred_probs": pred_probs,
    "use_softmax": False,
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
str_clean Ran
ran; our contract draft was wrong, not the code
allenai/easy-to-hard-generalization/utils/LM_utils.py
code served (permissive licence) · get_code("fd7c4aa91d08bfa9")
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 str. 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

ARGS = {
    "data": "  Hello World  "
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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_method_learning_curves Not yet run
unverified (harvested, no recorded run)
allenai/easy-to-hard-generalization/utils/plotting_utils.py
code served (permissive licence) · get_code("d2f07061f4f822d1")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 0 later) and 12 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
p_value 1 other paper
1 dated earlier
2111.13654 (2021-11)
Do Language Models Have Beliefs? Methods for Detecting, Updating, and Visualizing Model Beliefs

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

How can we train models to perform well on hard test data when hard training data is by definition difficult to label correctly? This question has been termed the scalable oversight problem and has drawn increasing attention as language models have continually improved. In this paper, we present the surprising conclusion that current pretrained language models often generalize relatively well from easy to hard data, even performing as well as oracle models finetuned on hard data. We demonstrate this kind of easy-to-hard generalization using simple finetuning methods like in-context learning, linear classifier heads, and QLoRA for seven different measures of datapoint hardness, including six empirically diverse human hardness measures (like grade level) and one model-based measure (loss-based). Furthermore, we show that even if one cares most about model performance on hard data, it can be better to collect easy data rather than hard data for finetuning, since hard data is generally noisier and costlier to collect. Our experiments use open models up to 70b in size and four publicly available question-answering datasets with questions ranging in difficulty from 3rd grade science questions to college level STEM questions and general-knowledge trivia. We conclude that easy-to-hard generalization in LMs is surprisingly strong for the tasks studied. 1 Test Input LM Generated Answer Q: John hires a driving service to get him to work each day. His work is 30 miles away and he has to go there and back each day. He goes to work 5 days a week for 50 weeks a year. He gets charged $2 per mile driven and he also gives his driver a $150 bonus per month How much does he pay a year for driving? A: John goes to work 5 days a week for 50 weeks a year. John goes to work 5 x 50 = <<5*50=250>>250 times a year. John pays $2 per mile driven. John pays $2 x 30 x 2 = <<2*30*2=120>>120 for each trip.

For agents

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

get_harvested_code_for_paper("2401.06751")
get_code_for_paper("2401.06751")
have("2401.06751")

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

curl https://syntology.ai/api/ran/2401.06751.json

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

[![Syntology run record](https://syntology.ai/api/ran/2401.06751.svg)](https://syntology.ai/paper/2401.06751)

Connect an agent — have() is free.