Skip to content
AI News HubLIVE
Original source3 min read

Tabular Foundation Models vs Tuned Gradient Boosting for Telecom Churn: A Profit- and Calibration-Aware Evaluation

Summary

This arXiv paper re-evaluates TabPFN models and TabICL v2 against Optuna-tuned gradient-boosted trees on five public telecom churn datasets, scoring expected maximum profit and calibration as well as ranking. TabPFN-3.5 earns significantly more profit on Cell2Cell and Iranian, but not on Orange; a larger tuning budget narrows the earlier 6.1-point PR-AUC gap to 4.0 points without closing it, and the TabPFN-3.5 weight licence, not scoring cost, is the main barrier to production.

SourcearXiv AIAuthor: Ismail Hossain Polas
Tabular Foundation Models vs Tuned Gradient Boosting for Telecom Churn: A Profit- and Calibration-Aware Evaluation
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 6 Nov 2026]

Title:Tabular Foundation Models vs Tuned Gradient Boosting for Telecom Churn: A Profit- and Calibration-Aware Evaluation

View a PDF of the paper titled Tabular Foundation Models vs Tuned Gradient Boosting for Telecom Churn: A Profit- and Calibration-Aware Evaluation, by Ismail Hossain Polas

View PDF HTML (experimental)

Abstract:Tabular foundation models (TFMs) predict from a labelled context in a single forward pass. A recent benchmark reported that they decisively outperform tree ensembles for customer churn, but it compared them with untuned gradient-boosted trees (GBDTs) and scored only ranking metrics, while a retention campaign is judged in money. We re-evaluate TabPFN-2.5, TabPFN-3.5, TabPFN-3.5-Fast and TabICL v2 against Optuna-tuned GBDTs on five public telecom churn datasets over 10 seeded splits. We separate three questions: whether a model ranks churners well, whether that ranking earns more expected maximum profit (EMP), and whether its probabilities can drive a threshold policy. With a split-aware test and Holm correction over 20 pre-declared comparisons, TabPFN-3.5 earns significantly more EMP than the best tuned GBDT on Cell2Cell (1.9%) and Iranian (7.6%), and these two gains also hold against each of the three tuned GBDTs separately. On Orange it is ahead in all 10 splits (14.3%), but the gain is not significant after correction (Holm p = 0.155). The splits resample the same datasets, so these results do not stand for independent operators. TabICL v2, the earlier benchmark's leader, beats the best tuned GBDT only on Iranian, not all three of them, and underpredicts Orange churn by 2.1 points on a 7.3% base rate. On the datasets and splits where we tested them, a substantially larger tuning budget and a duplicate-aware split leave the main conclusions unchanged. Retuning the baseline cuts the earlier headline gap from 6.1 to 4.0 PR-AUC points without closing it. The licence of the TabPFN-3.5 weights we used (License v1.0), not scoring cost, is what stands between its gains and production use.

Comments: 17 pages, 2 figures, 14 tables. Code, per-split predictions and reproducibility audit: this https URL (archived at this https URL)

Subjects:

Artificial Intelligence (cs.AI); Machine Learning (cs.LG)

ACM classes: I.2.6; H.2.8

Cite as: arXiv:2611.12942 [cs.AI]

(or arXiv:2611.12942v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2611.12942

arXiv-issued DOI via DataCite

Submission history

From: Ismail Hossain Polas [view email] [v1] Fri, 6 Nov 2026 17:15:06 UTC (49 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Tabular Foundation Models vs Tuned Gradient Boosting for Telecom Churn: A Profit- and Calibration-Aware Evaluation, by Ismail Hossain Polas

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.AI

new | recent | 2026-11

Change to browse by:

cs cs.LG

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • The study compares TabPFN-2.5, TabPFN-3.5, TabPFN-3.5-Fast and TabICL v2 with Optuna-tuned GBDTs across five telecom churn datasets and 10 seeded splits.
  • TabPFN-3.5 shows significant expected maximum profit gains on Cell2Cell (1.9%) and Iranian (7.6%), and also against each tuned GBDT separately; Orange's 14.3% gain is not significant after Holm correction (p = 0.155).
  • Retuning the baseline narrows the earlier PR-AUC gap from 6.1 to 4.0 points without eliminating it.
  • The License v1.0 terms on the TabPFN-3.5 weights, rather than scoring cost, are what blocks production use.

Highlights and analysis are generated automatically and may contain errors. Check the original source.