ACL Anthology · ACL 2026:acl:2026.acl-demo.1 · 记录 #1
Interpreto: An Explainability Library for Transformers
Antonin Poché Institut de Recherche en Informatique de Toulouse and IRT Saint ExuperyThomas Mullor INSTITUTE OF TECHNOLOGICAL RESEARCH IRT SAINT EXUPERYGabriele SartiFrédéric Boisnard INSTITUTE OF TECHNOLOGICAL RESEARCH IRT SAINT EXUPERYCorentin Friedrich IRT Saint ExupéryCharlotte Claye Centrale SupélecFrancois Hoofd IRT SAINT EXUPERYRaphael BernasNicholas Asher CNRSCeline Hudelot CentraleSupelecFanny Jourdan IRT Saint Exupery
来源日期（仅到月份） 2026-07
来源记录日期 2026-07
雷达观测 2026-11-01 22:10
摘要
Interpreto is an open-source Python library for interpreting HuggingFace language models, from early BERT variants to LLMs. It provides two complementary families of methods: attribution methods and concept-based explanations. The library bridges recent research and practical tooling by exposing explanation workflows through a unified API for both classification and text generation. A key differentiator is its end-to-end concept-based pipeline (from activation extraction to concept learning, interpretation, and scoring), which goes beyond feature-level attributions and is uncommon in existing libraries.