SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.26017 · 2026

QUITOBENCH: A High-Quality Open Time Series Forecasting Benchmark

Fan Zhou, Rui Wang, Wei Zhang, Hang Yu, Siqiao Xue, Jianguo Li, Peng Di, Zhaoyang Zhu, Rongyao Cai, Yixiang Mu

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 0 have no contract verdict; 2 failures of our instrument; 0 no recorded run), 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/2603.26017.json · badge: /api/ran/2603.26017.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
alipay/quito canonical (classifier label)
classifier basis: census
1 ran · 0 no recorded run
SerendipityOneInc/look-bench canonical (classifier label)
classifier basis: census
1 ran · 0 no recorded run
FunctionStatusWhere it lives
MSE Ran
ran; our contract draft was wrong, not the code
alipay/quito/quito/metrics.py
code served (permissive licence) · get_code("2136500bcdbd55d3")
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 torch
import torch.nn as nn

torch.manual_seed(0)

# Create small tensors for y_pred and y_true
# Batch size: 2, Feature dimension: 4
y_pred = torch.randn(2, 4)
y_true = torch.randn(2, 4)

ARGS = {
    "y_pred": y_pred,
    "y_true": y_true,
}
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
load_zooclaw_task Ran
ran; our contract draft was wrong, not the code
SerendipityOneInc/look-bench/datasets/zooclaw_dataset.py
code served (permissive licence) · get_code("548f7e5fd955961b")
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 os
import tempfile
from pathlib import Path
from PIL import Image
import torch
from torchvision import transforms

torch.manual_seed(0)

# Create a temporary directory structure for the task
temp_dir = tempfile.mkdtemp()
task_dir = Path(temp_dir) / "zooclaw-fashion" / "text2image"
task_dir.mkdir(parents=True, exist_ok=True)

# Create sample query data
queries = [
    {"query_id": 1, "short_query": "red dress", "long_query_llm": "A beautiful long red evening dress with flowing fabric"},
    {"query_id": 2, "short_query": "blue shoes", "long_query_llm": "Blue leather shoes with comfortable sole"}
]

# Create sample corpus data
corpus = [
    {"corpus_id": 10, "image_path": "images/item1.jpg"},
    {"corpus_id": 11, "image_path": "images/item2.jpg"},
    {"corpus_id": 12, "image_path": "images/item3.jpg"}
]

# Create sample ground truth
ground_truth = {
    "1": [10, 11],
    "2": [11, 12]
}

# Create sample queries_long.json
queries_long = [
    {"query_id": 1, "short_query": "red dress", "long_query_llm": "A beautiful long red evening dress with flowing fabric"},
    {"query_id": 2, "short_query": "blue shoes", "long_query_llm": "Blue leather shoes with comfortable sole"}
]

# Write JSON files
with open(task_dir / "queries.json", "w") as f:
    json.dump(queries, f)

with open(task_dir / "queries_long.json", "w") as f:
    json.dump(queries_long, f)

with open(task_dir / "corpus.json", "w") as f:
    json.dump(corpus, f)

with open(task_dir / "ground_truth.json", "w") as f:
    json.dump(ground_truth, f)

# Create sample image files
images_dir = task_dir / "images"
images_dir.mkdir(exist_ok=True)

for i in range(1, 4):
    img = Image.new('RGB', (64, 64), color=(i*50, i*60, i*70))
    img.save(images_dir / f"item{i}.jpg")

# Define image transform
image_transform = transforms.Compose([
    transforms.Resize((32, 32)),
    transforms.ToTensor(),
])

# Construct ARGS
ARGS = {
    "task_dir": str(task_dir),
    "task_type": "text2image",
    "transform": image_transform,
    "data_root": str(task_dir),
    "use_long_queries": False
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
PIL, torch, torchvision — 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

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

Time series forecasting is critical across finance, healthcare, and cloud computing, yet progress is constrained by a fundamental bottleneck: the scarcity of large-scale, high-quality benchmarks. To address this gap, we introduce QUITO-BENCH, a regime-balanced benchmark for time series forecasting with coverage across eight trend×seasonality×forecastability (TSF) regimes, designed to capture forecasting-relevant properties rather than application-defined domain labels. The benchmark is built upon QUITO, a billion-scale time series corpus of application traffic from Alipay spanning nine business domains. Benchmarking 10 models from deep learning, foundation models, and statistical baselines across 232,200 evaluation instances, we report four key findings: ① a context-length crossover where deep learning models lead at short context (L = 96) but foundation models dominate at long context (L ≥ 576); ② forecastability is the dominant difficulty driver, producing a 3.64× MAE gap across regimes; ③ deep learning models match or surpass foundation models at 59× fewer parameters; and ④ scaling the amount of training data provides substantially greater benefit than scaling model size for both model families. These findings are validated by strong cross-benchmark and cross-metric consistency. Our open-source release enables reproducible, regime-aware evaluation for time series forecasting research.

For agents

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

get_harvested_code_for_paper("2603.26017")
get_code_for_paper("2603.26017")
have("2603.26017")

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

curl https://syntology.ai/api/ran/2603.26017.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.26017.svg)](https://syntology.ai/paper/2603.26017)

Connect an agent — have() is free.