anhaidgroup/polaris-arctic-v2

Arctic v2

Arctic is one of six datasets in Polaris: Learning to Generate Table Descriptions from Retrieval Feedback, alongside aw, lter, ecir, wikitables, and wtr.

It holds 251 tables sampled at random from the Environmental Data Initiative (EDI), a repository of long-term ecological research data — lake water temperature, soil chemistry, rainfall, coral taxonomy — and 20 keyword queries over them. For each query–table pair, a person decided whether that table answers that query; those decisions are the relevance judgments, and they live in qrels.csv.

Each Polaris dataset comes in two versions. v2, this one, adds the tuples — the rows of each table, in tuples.zip — on top of the table metadata, the queries, and the relevance judgments. v1 is the same without the tuples and is at polaris-arctic-v1.

Files

queries.csv — one row per query.

query_id,query
q17,nitrous oxide
q8,average soil volumetric water content data: 2013 - ongoing

qrels.csv — one row per relevant pair, here the three tables relevant to q17.

query_id,table_id,relevance_score
q17,115,1
q17,185,1
q17,248,1

Judging was done by pooling: for each query, several retrievers proposed candidate tables and a person labelled each one, 1 for relevant and 0 for not. Only the 1s are listed here, so a table missing for a query counts as not relevant.

metadata.csv — one row per table, here the same three tables with their column lists cut short.

table_id,table_name,column_names
115,HBEF Ice Storm Trace Gas Data,"[""Project"", ""Year"", ..., ""CO2"", ""N2O"", ""CH4""]"
185,VCR09168_1,"[""YEAR"", ""DOY_JULIAN"", ""CO2_FLUX"", ""N2O_FLUX"", ...]"
248,subtgflx.pb.data.csv,"[""sample_date"", ""loc_code"", ""tmt_code"", ""N2O_flux"", ""CO2_flux""]"

tuples.zip — one CSV per table, at Tuples/<table_id>.csv. 115 is HBEF Ice Storm Trace Gas Data, one of the three tables relevant to q17. Its first rows:

Project,Year,Month,TrtYr,Season,Treatment,Plot,Chamber,CO2,N2O,CH4
ISE,2015,9,0,F,MIDx2,1,1,0.14,5.56,-1.83
ISE,2015,9,0,F,MIDx2,1,4,0.07,-2.76,-2.72
ISE,2015,9,0,F,MIDx2,1,5,0.1,0.93,-3.39

Each file's header matches that table's column_names. Tables were capped at 500,000 rows; 9 of the 251 hit that cap.

Statistics

A table counts as gold if it scores above 0 for at least one query.

Statistic	Value
Domain	Science
Tables	251
Queries	20
Gold tables	121
Relevant tables per query	min 3, max 43, average 18.5
Metadata fields	table name, column names
Relevance	binary
Tuples	15,528,869 rows across the tables
Rows per table	median 979, min 1, max 500,000
tuples.zip	137 MB, 1.0 GB unzipped

Download

The repo is about 137 MB, almost all of it tuples.zip, which unpacks to 1.0 GB. You do not need a Hugging Face account to download it.

Option 1 — click the Files tab at the top of this page and save each file.

Option 2 — command line (recommended):

pip install huggingface_hub
hf download anhaidgroup/polaris-arctic-v2 --repo-type dataset --local-dir arctic

Polaris has six datasets in total and Arctic is one of them. Each sits in its own repository, so to download all six quickly — the v2 repositories, with tuples, about 4.8 GB in total:

for d in aw arctic lter ecir wikitables wtr; do
  hf download anhaidgroup/polaris-$d-v2 --repo-type dataset --local-dir polaris_v2/$d
done

Usage

column_names is an array, so it needs parsing when you load the file. For example:

import ast
import pandas as pd

metadata = pd.read_csv("arctic/metadata.csv")
queries = pd.read_csv("arctic/queries.csv")
qrels = pd.read_csv("arctic/qrels.csv")

metadata["columns"] = metadata["column_names"].apply(ast.literal_eval)

relevant = qrels.loc[qrels.query_id == "q17", "table_id"].tolist()
# [115, 185, 248]

To read one table's tuples without unpacking the archive:

import zipfile

with zipfile.ZipFile("arctic/tuples.zip") as z:
    with z.open("Tuples/115.csv") as f:
        rows = pd.read_csv(f)

How is this dataset created?

The tables are a random sample of 251 tables from the Environmental Data Initiative, a scientific data repository that stores long-term ecological and environmental research data.

The Polaris authors wrote 20 queries covering topics such as precipitation, species diversity, and nitrous oxide. They then manually labelled the table–query pairs, giving a binary score of 0 or 1.

Citation

@misc{cai2026polaris,
  title         = {Polaris: Learning to Generate Table Descriptions from Retrieval Feedback},
  author        = {Cai, Ting and Phan, Tuan Minh and Doan, AnHai},
  year          = {2026},
  eprint        = {2608.17171},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CL},
  doi           = {10.48550/arXiv.2608.17171},
  url           = {https://arxiv.org/abs/2608.17171}
}

Please also cite the source of the tables:

@misc{edi,
  author = {Paul Hanson},
  title  = {Environmental Data Initiative},
  year   = {2025},
  url    = {https://edirepository.org},
}

License

What	Comes from	License
Table metadata and tuples	EDI, uploaded by the researchers who collected the data	CC0 1.0 by default, though researchers can set stricter terms on their own data
Queries and relevance judgments	The Polaris authors	ODC-BY 1.0

Contact

Email minhrua@cs.wisc.edu, valid until May 2029. After that, email anhai@cs.wisc.edu.