Statistical Deficiency for Task Inclusion Estimation
Loïc Fosse, Frederic Bechet, Benoit Favre, Géraldine Damnati, Gwénolé Lecorvé, Maxime Darrin, Philippe Formont, Pablo Piantanida
Abstract
Tasks are central in machine learning, as they are the most natural objects to assess the capabilities of current models. The trend is to build general models able to address any task. Even though transfer learning and multitask learning try to leverage the underlying task space, no well-founded tools are available to study its structure. This study proposes a theoretically grounded setup to define the notion of task and to compute the inclusion between two tasks from a statistical deficiency point of view. We propose a tractable proxy as information sufficiency to estimate the degree of inclusion between tasks, show its soundness on synthetic data, and use it to reconstruct empirically the classic NLP pipeline.
BibTeX
@inproceedings{fosse-etal-2025-statistical,
title = "Statistical Deficiency for Task Inclusion Estimation",
author = {Fosse, Lo{\"i}c and
Bechet, Frederic and
Favre, Benoit and
Damnati, G{\'e}raldine and
Lecorv{\'e}, Gw{\'e}nol{\'e} and
Darrin, Maxime and
Formont, Philippe and
Piantanida, Pablo},
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.18/",
doi = "10.18653/v1/2025.acl-long.18",
pages = "382--415",
ISBN = "979-8-89176-251-0"
}