SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2601.19040 · 2026

OATS: Online Data Augmentation for Time Series Foundation Models

Jiaqi Ma, Chang Xu, Jiang Bian, Li Zhao, Junwei Deng, Ming Jin, Xu Zhang, Chenghao Liu

arXiv · PDF · Open in the Atlas

Code that ran

11 of 12 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; 4 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 5 failures of our instrument; 1 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, measured 2026-11-22; 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/2601.19040.json · badge: /api/ran/2601.19040.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
microsoft/TimeCraft canonical (classifier label)
classifier basis: census
11 ran · 1 no recorded run
FunctionStatusWhere it lives
identity Ran
ran, honoured the contract we drafted
microsoft/TimeCraft/DiGA/diffusion/cont_ctrl_net.py
code served (permissive licence) · get_code("7f1040f5e3991d5e")
How it ran

Ran in Syntology's jail on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) 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. The comparison behind this sample's level is a different run: the peer-agreement lane ran it again on the host, not in the jail (report dated 2026-09-05T00:37:21Z). 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)

# The identity function simply returns its first argument
# and ignores all other args and kwargs
# So we need to provide a value for t and can optionally provide other args/kwargs

t = torch.randn(2, 4)
extra_arg_1 = torch.randn(3, 5)
extra_arg_2 = "some_string"
extra_kwarg_1 = 42
extra_kwarg_2 = [1, 2, 3]

ARGS = {
    "t": t,
    "args": (extra_arg_1, extra_arg_2),
    "kwargs": {"key1": extra_kwarg_1, "key2": extra_kwarg_2}
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
default Ran
ran, violated the contract we drafted
microsoft/TimeCraft/DiGA/diffusion/cont_ctrl_net.py
code served (permissive licence) · get_code("d1ef6b8cb9a28a53")
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 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

torch.manual_seed(0)

def exists(x):
    return x is not None

def default(val, d):
    if exists(val):
        return val
    return d() if callable(d) else d

# Test case 1: val is not None, should return val
val_1 = "test_value"
d_1 = lambda: "default_value"

# Test case 2: val is None, d is callable, should return d()
val_2 = None
d_2 = lambda: "computed_default"

# Test case 3: val is None, d is not callable, should return d
val_3 = None
d_3 = "static_default"

# Test case 4: val is a number
val_4 = 42
d_4 = 100

# Test case 5: val is a tensor
val_5 = torch.tensor([1.0, 2.0, 3.0])
d_5 = lambda: torch.tensor([0.0])

ARGS = {
    "val": val_1,
    "d": d_1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
exists Ran
ran, violated the contract we drafted
microsoft/TimeCraft/DiGA/diffusion/cont_ctrl_net.py
code served (permissive licence) · get_code("608e364a9d2376a3")
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 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

def exists(x):
    return x is not None

ARGS = {
    "x": torch.tensor([1, 2, 3, 4])
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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
generate_dataset_description Ran
ran; our contract draft was wrong, not the code
microsoft/TimeCraft/BRIDGE/ts_to_text.py
code served (permissive licence) · get_code("d7c15dd02c7f9d9f")
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 random
import json
import tempfile
import os

# Create a temporary JSON file with dataset metadata
temp_dir = tempfile.mkdtemp()
json_file_path = os.path.join(temp_dir, 'dataset_description_bank.json')

dataset_templates = [
    "Dataset: {dataset_name} | Frequency: {frequency} | Description: {data_description} | Period: {start_date} to {end_date} | Domain: {domain}",
    "{dataset_name} - A {domain} dataset with {frequency} frequency. Data: {data_description}. Time span: {start_date}-{end_date}",
]

datasets = {
    "stock_prices": {
        "frequency": "daily",
        "data_description": "Historical stock price data including OHLC values",
        "start_date": "2020-01-01",
        "end_date": "2023-12-31",
        "domain": "finance"
    },
    "weather_data": {
        "frequency": "hourly",
        "data_description": "Temperature, humidity, and precipitation measurements",
        "start_date": "2022-01-01",
        "end_date": "2023-12-31",
        "domain": "meteorology"
    }
}

with open(json_file_path, 'w') as f:
    json.dump(datasets, f)

# Make dataset_templates available globally as the function expects it
import sys
current_module = sys.modules[__name__]
setattr(current_module, 'dataset_templates', dataset_templates)

ARGS = {
    'dataset_name': 'stock_prices',
    'json_file': json_file_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 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_logger Ran
ran; our contract draft was wrong, not the code
microsoft/TimeCraft/Diff-MN/lib/utils.py
code served (permissive licence) · get_code("6e6961df228cbc5a")
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 RootLogger. 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 logging
import tempfile
import os

# Create temporary files for testing
temp_dir = tempfile.mkdtemp()
logpath = os.path.join(temp_dir, "test.log")
filepath = "main_script.py"

# Create a temporary package file to read
package_file = os.path.join(temp_dir, "package_module.py")
with open(package_file, 'w') as f:
    f.write("# This is a test package file\ndef hello():\n    pass\n")

ARGS = {
    'logpath': logpath,
    'filepath': filepath,
    'package_files': [package_file],
    'displaying': True,
    'saving': True,
    'debug': False
}
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 RootLogger — a RootLogger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a RootLogger 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_pseudo_price Ran
ran, honoured the contract we drafted
microsoft/TimeCraft/DiGA/generate.py
code served (permissive licence) · get_code("74cc4fa33614920f")
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. 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

class Args:
    def __init__(self, random_price=False, pseudo_price=1.5):
        self.random_price = random_price
        self.pseudo_price = pseudo_price

# Test case 1: with random_price=True
args_random = Args(random_price=True, pseudo_price=1.5)

# Test case 2: with random_price=False
args_fixed = Args(random_price=False, pseudo_price=2.0)

ARGS = {
    "args": args_fixed
}
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 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
load_json_file Ran
ran; our contract draft was wrong, not the code
microsoft/TimeCraft/BRIDGE/ts_to_text.py
code served (permissive licence) · get_code("a284824bddec284c")
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

# Create a temporary JSON file with valid content
temp_dir = tempfile.gettempdir()
temp_filepath = os.path.join(temp_dir, "test_load_json.json")

# Write a valid JSON file
test_data = {
    "key1": "value1",
    "key2": [1, 2, 3],
    "key3": {"nested": "object"}
}

with open(temp_filepath, 'w', encoding='utf-8') as f:
    json.dump(test_data, f)

ARGS = {
    "filepath": temp_filepath
}
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
load_pickle Ran
ran; our contract draft was wrong, not the code
microsoft/TimeCraft/Diff-MN/lib/utils.py
code served (permissive licence) · get_code("93361336a4f19653")
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 pickle
import tempfile
import os

# Create a temporary pickle file with some test data
test_data = {
    'key1': 'value1',
    'key2': [1, 2, 3],
    'key3': {'nested': 'dict'}
}

# Create a temporary file and write pickle data to it
temp_fd, temp_path = tempfile.mkstemp(suffix='.pkl')
os.close(temp_fd)

with open(temp_path, 'wb') as f:
    pickle.dump(test_data, f)

# Define ARGS with the filename parameter
ARGS = {
    'filename': temp_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
load_text_file Ran
ran; our contract draft was wrong, not the code
microsoft/TimeCraft/BRIDGE/ts_to_text.py
code served (permissive licence) · get_code("31a9092f1b396c58")
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 tempfile
import os

# Create a temporary text file with some content
temp_dir = tempfile.mkdtemp()
filepath = os.path.join(temp_dir, "test.txt")

with open(filepath, 'w', encoding='utf-8') as f:
    f.write("Hello, World!\nThis is a test file.\n")

ARGS = {
    "filepath": filepath
}
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 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
reparameterize Ran
ran, honoured the contract we drafted
microsoft/TimeCraft/Diff-MN/models/kovae.py
pointer only (licence: MIT); the stored copy get_code serves, from azencot-group/KoVAE, is pointer only (licence: NONE) · get_code("dadec7f7cc09688a")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from azencot-group/KoVAE, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

safe_convert_to_float Ran
ran, honoured the contract we drafted
microsoft/TimeCraft/BRIDGE/llm4text2ts.py
code served (permissive licence) · get_code("ccdaa08ef4db0396")
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 ast

def safe_convert_to_float(series):
    if isinstance(series, str):
        try:
            series = ast.literal_eval(series)
        except (ValueError, SyntaxError) as e:
            print(f'Error converting series: {series}, error: {e}')
            return None
    if isinstance(series, (list, tuple)):
        return [el if isinstance(el, (int, float, list, tuple)) else None for el in series]
    elif isinstance(series, (int, float)):
        return series
    else:
        print(f'Unexpected type for series: {series}, type: {type(series)}')
        return None


# Test case 1: string representation of a list
test_series_1 = "[1, 2.5, 3]"

# Test case 2: direct list with mixed types
test_series_2 = [1, 2.5, "invalid", 4]

# Test case 3: direct float
test_series_3 = 3.14

# Test case 4: direct int
test_series_4 = 42

# Test case 5: tuple
test_series_5 = (1.0, 2.0, 3.0)

# Test case 6: string with tuple representation
test_series_6 = "(10, 20.5, 30)"

ARGS = {
    "series": test_series_1
}
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 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
make_dataset Not yet run
unverified (harvested, no recorded run)
microsoft/TimeCraft/Diff-MN/lib/utils.py
code served (permissive licence) · get_code("03cd51541800f514")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 1 the same month, 4 later) and 4 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
exists 144 other papers
135 dated earlier, 2 the same month, 7 later · 40 only as an assembly brick
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1711.11585 (2017-11)
High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
and 141 more not named here
default 79 other papers
73 dated earlier, 6 later · 15 only as an assembly brick
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
2010.02502 (2020-10)
Denoising Diffusion Implicit Models
and 76 more not named here
identity 54 other papers
50 dated earlier, 4 later · 11 only as an assembly brick
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1907.01470 (2019-07)
Augmenting Self-attention with Persistent Memory
2006.11239 (2020-06)
Denoising Diffusion Probabilistic Models
and 51 more not named here
reparameterize 3 other papers
2 dated earlier, 1 later
2305.15924 (2023-05)
Sample and Predict Your Latent: Modality-free Sequential Disentanglement via Contrastive Estimation
2310.02619 (2023-10)
Generative Modeling of Regular and Irregular Time Series Data via Koopman VAEs
2605.17804 (2026-05)
GenTS: A Comprehensive Benchmark Library for Generative Time Series Models
get_logger 2 other papers
2 dated earlier
2106.13898 (2021-06)
Closed-form Continuous-time Neural Models
2204.09633 (2022-04)
SurvLatent ODE : A Neural ODE based time-to-event model with competing risks for longitudinal data improves cancer-associated Venous Thromboembolism (VTE) prediction
load_pickle 2 other papers
2 dated earlier
2106.13898 (2021-06)
Closed-form Continuous-time Neural Models
2204.09633 (2022-04)
SurvLatent ODE : A Neural ODE based time-to-event model with competing risks for longitudinal data improves cancer-associated Venous Thromboembolism (VTE) prediction
make_dataset 2 other papers
2 dated earlier
2106.13898 (2021-06)
Closed-form Continuous-time Neural Models
2204.09633 (2022-04)
SurvLatent ODE : A Neural ODE based time-to-event model with competing risks for longitudinal data improves cancer-associated Venous Thromboembolism (VTE) prediction
load_json_file 1 other paper
1 dated earlier
2510.17633 (2025-10)
SARSteer: Safeguarding Large Audio-Language Models via Safe-Ablated Refusal Steering

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

Time Series Foundation Models (TSFMs) are a powerful paradigm for time series analysis and are often enhanced by synthetic data augmentation to improve the training data quality. Existing augmentation methods, however, typically rely on heuristics and static paradigms. Motivated by dynamic data optimization, which shows that the contribution of samples varies across training stages, we propose OATS (Online Data Augmentation for Time Series Foundation Models), a principled strategy that generates synthetic data tailored to different training steps. OATS leverages valuable training samples as principled guiding signals and dynamically generates high-quality synthetic data conditioned on them. We further design a diffusion-based framework to produce realistic time series and introduce an explore-exploit mechanism to balance efficiency and effectiveness. Experiments on TSFMs demonstrate that OATS consistently outperforms regular training and yields substantial performance gains over static data augmentation baselines across six datasets and two TSFM architectures. The code is available at the link https://github.com/microsoft/TimeCraft.

For agents

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

get_harvested_code_for_paper("2601.19040")
get_code_for_paper("2601.19040")
have("2601.19040")

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

curl https://syntology.ai/api/ran/2601.19040.json

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

[![Syntology run record](https://syntology.ai/api/ran/2601.19040.svg)](https://syntology.ai/paper/2601.19040)

Connect an agent — have() is free.