Unfair Clause Detection

Seven large language models, prompted in three ways, face fine-tuned classifiers at spotting unfair clauses in Terms of Service, and the fine-tuned classifiers still win.

With good machine translation, an English unfair-clause detector for Terms of Service matches detectors trained on new German, Italian, and Polish corpora.

Memory networks read a knowledge base of expert legal rationales to detect unfair clauses in terms of service and to point to the reasons.

Annotations made by legal experts on English contracts are projected onto their German versions, even when the two versions do not match sentence by sentence.

CLAUDETTE now tells consumers why a Terms of Service clause may be unfair, by linking each flagged clause to legal rationales written by experts.

CLAUDETTE uses machine learning to detect potentially unfair clauses in online terms of service, and could serve lawyers and consumers alike.

Linking unfair clauses in Terms of Service to a knowledge base of rationales may let memory networks explain their outputs and detect unfair clauses better.